An intelligent perception-based asphalt pavement semi-rigid base crack monitoring method

Abstract

The application discloses an intelligent sensing-based asphalt pavement semi-rigid base crack monitoring method, which is characterized in that: an electric resistance sensor integrated intelligent particle is buried in the asphalt pavement semi-rigid base to collect the mechanical response of the pavement structure, and the base crack is monitored. The steps are as follows: based on the conversion relationship of the macroscopic and microscopic parameters calibrated by the uniaxial tension and compression simulation experiment, a cross-scale pavement model is established by adopting a finite element-discrete element coupling method; through numerical simulation of the base crack evolution, the base crack is tracked and positioned by combining the Westergard complex variable function method of the I-II type composite crack of fracture mechanics; the intelligent particle is buried in the base of the pavement model to collect information, and according to the stress concentration characteristics of the crack neighborhood, the matching of the intelligent particle monitoring index and the extraction of the sensitive range are realized; the intelligent particle laying scheme for the semi-rigid base crack monitoring and positioning is established, and the monitoring of the semi-rigid base crack position and the expansion stage is realized.

Description

Technical Field

[0001] This invention relates to the field of intelligent monitoring technology for the internal structure of intelligent roads and pavements, specifically a method for multi-scale calculation and location monitoring of cracks in the semi-rigid base course of asphalt pavements based on intelligent particle sensing. Background Technology

[0002] Semi-rigid base asphalt pavement is the main type of pavement structure for high-grade highways in my country. However, with the continuous use of semi-rigid base pavement structures, a large number of cracks have appeared in the semi-rigid base, seriously affecting the pavement's performance and service life, becoming a major defect of this pavement structure. The timing and location of crack appearance in semi-rigid bases during pavement construction and use are currently not definitively established in the industry. Therefore, timely understanding and tracking of the initiation and propagation process of cracks in semi-rigid bases can provide a scientific basis for subsequent pavement maintenance and treatment, allowing for timely and effective intervention in the development of cracks in the semi-rigid base, thereby improving pavement performance and preventing further deterioration.

[0003] The performance of asphalt pavement is closely related to its internal mechanical response. Road researchers primarily rely on numerical simulations, using finite element or discrete element methods to calculate the internal mechanical properties of materials and study road performance and damage evolution. Simultaneously, researchers also implant sensors within the pavement structure of actual engineering test roads to dynamically collect key mechanical parameters in real time, comparing them with the mechanical properties obtained from numerical simulations. This allows for analysis of the mechanical behavior within the actual pavement structure and study of the evolution of asphalt pavement service performance. However, due to the complex mechanical properties of semi-rigid base materials, there is limited research in the industry on sensing and analyzing the mechanical response of semi-rigid base materials, and the propagation process of cracks in the base layer cannot be monitored.

[0004] Intelligent particles are particle-type high-temperature resistant sensors developed to meet the monitoring needs of railway subgrades and highway pavements. Internally, they contain sensing elements such as a triaxial gyroscope, triaxial accelerometer, and triaxial magnetometer, enabling them to monitor changes in parameters such as particle rotation angle, displacement, stress magnitude, and temperature. They utilize low-power Bluetooth and other communication protocols to acquire pavement signals and store them in the cloud in real time. They are convenient to use and unaffected by environmental conditions. Therefore, this invention proposes a method for monitoring cracks in semi-rigid base courses based on this monitoring approach. Summary of the Invention

[0005] Technical problem to be solved: This invention addresses the problem of numerous cracks appearing in existing semi-rigid base courses, which seriously affect the pavement's performance and service life, becoming a major structural defect. Furthermore, the complex mechanical properties of semi-rigid base course materials have resulted in limited research on the perception and analysis of their mechanical response, and the inability to monitor the propagation process of cracks in the base course. This invention provides a method for monitoring cracks in semi-rigid base courses of asphalt pavements based on intelligent sensing.

[0006] Technical solution:

[0007] A method for monitoring cracks in semi-rigid base courses of asphalt pavements based on intelligent sensing includes the following steps:

[0008] Step 1: Through matching uniaxial tension and compression numerical simulation experiments, the transformation relationship between macroscopic and microscopic parameters is obtained;

[0009] Step 2: Establish a multi-scale semi-rigid base pavement model using the finite element-discrete element coupling method;

[0010] Step 3: Through numerical simulation of the evolution of cracks in semi-rigid base layers, combined with the Westergard complex variable function method for fracture mechanics type I-II composite cracks, the propagation of cracks in composite semi-rigid base layers is tracked and located.

[0011] Step 4: Smart particles are embedded in the semi-rigid base of the road model to collect information. Based on the response characteristics of stress concentration in the vicinity of cracks in the semi-rigid base, the monitoring indicators of smart particles and the sensitive range are matched and extracted during the crack propagation process.

[0012] Step 5: Establish an intelligent particle layout density scheme for monitoring and locating cracks in semi-rigid base layers, so as to realize the monitoring of the location and expansion stage of cracks in semi-rigid base layers.

[0013] As a preferred embodiment of the present invention: the first step specifically involves: firstly, establishing a cubic model of the same size with a 30*30*30 finite element mesh and 30*30*30 centripetally arranged discrete element particles, setting the constitutive relation of the linear elastic model, and simultaneously assigning the same density ρ = 2500 kg / m³. 3 The model was designed with an elastic modulus E = 105.20 GPa and a Poisson's ratio ν = 0.314 to control the same uniaxial tension and compression conditions and properties. The same number of meshes and particles were used to achieve macro-microscopic comparison. Next, the bottom surface of the model was fixed, while the sides were unrestricted. A single stress control mode of 1 MPa and a single strain control mode of 1% were applied to the top surface. A stress mode with a period of 10 Hz and a magnitude of 1 MPa, and a strain mode with a period of 10 Hz and a magnitude of 1% were also applied. The overall stress and strain state of the model was extracted. The correspondence between the microscopic parameters of the particles in the discrete element method and the macroscopic parameters of the elements in the finite element method was obtained by matching the following formula:

[0014]

[0015]

[0016]

[0017]

[0018]

[0019] Among them, K n For the normal stiffness between particles, K s For shear stiffness, X b For the fracture displacement, F s0 For shear resistance, μ p μ is the coefficient of friction between particles. i C is the inherent coefficient of friction. u T represents uniaxial compressive strength. u Where d is the tensile strength and d is the particle diameter;

[0020] Next, extract the displacement u of each element in the finite element model under load. x u y u z , element stress σ x σ y σ z Extract the particle contact force, displacement, and rotation angle from the discrete element model, and calculate the displacement of each particle as u. ′ x =X nxi +X sxi ,

[0021] ′

[0022] u y =X nyi +X syi ,

[0023] u ′ z =X nzi +X szi , where X nxi X sxi X nyi X syi X nzi X szi

[0024] Let x, y, and z be the x, y, and z components of the relative displacement between the normal and tangential directions of the i-th particle, and let σ be the particle contact stress. ′ x =(Fnxi +F sxi ) / S x σ ′ y =(F nyi +F syi ) / S y σ ′ z =(F nzi +F szi ) / S z F nxi F sxi F nyi F syi F nzi F szi These represent the x, y, and z components of the normal and tangential contact forces of the particles, respectively; S x S y S z These represent the contact areas between particles in the x, y, and z directions, respectively.

[0025] Let the strain u be obtained by fitting the particle model. ′ i Stress σ ′ i respectively satisfy f(u) ′ i )=c0+c1i+c2i 2 g(σ) i i )=d0+d1i+d2i 2 ,set up Where c0, c1, c2, d0, d1, and d2 are constants, calculate the inner product and substitute it into the normal equation system:

[0026]

[0027]

[0028] Solve for f(u) ′ i )=c0+c1i+c2i 2 g(σ) ′ i )=d0+d1i+d2i 2 Fit u ′ i and σ ′ i Curve; then calculate the standardized Euclidean distance using the following formula:

[0029]

[0030]

[0031] Among them, s, s ′ The standard deviation of particle displacement and stress data; finally, the displacement values ​​at each point in the finite element method are corrected to u. ′ =u±d(u), stress value corrected to σ ′ =σ±d(σ), then the correction value of the elastic modulus of the finite element model. Poisson's ratio correction value v ′ =v±2d(u), the macro-micro transformation relationship is corrected as follows:

[0032]

[0033]

[0034]

[0035]

[0036]

[0037] As a preferred technical solution of the present invention: the second step specifically involves establishing a road structure with an overall size of 8*8*8m, ​​each surface layer being 4+6+8cm and the semi-rigid base layer being 60cm. The surface layer and the semi-rigid base layer are set as finite element elastic element mesh models, and the foundation is set as a finite element Mohr-Coulomb mesh model. A discrete element region of size 60*60*60cm is selected in the middle of the semi-rigid base layer and filled with particles of different sizes to simulate heterogeneous aggregates. 1mm particles are filled to simulate cement binder. Based on the transformation relationship between macroscopic and microscopic material parameters, the material parameters of the finite element region and the discrete element region of the semi-rigid base layer are calibrated.

[0038] As a preferred technical solution of the present invention: the third step specifically involves approximating the stress field range where stress concentration occurs in the crack neighborhood as an irregular column stretched from a spherical cubic region along the crack propagation direction; changing the load conditions in the road surface numerical model, and extracting the mechanical response of the semi-rigid base layer in the crack tip neighborhood when the semi-rigid base layer is damaged after cracks appear, to obtain the column and spherical radius R at each stage and in each direction. i Stress threshold point size σ Ci The average values ​​are used to obtain R0 and σ. C ;

[0039] Based on the Westergard complex variable function method for type I-II composite cracks in fracture mechanics, the stress and displacement values ​​at the above points are calculated using the following formula: i θ i :

[0040]

[0041]

[0042] Where, σ x σ y τ xy These represent the stress in the x-direction, stress in the y-direction, and shear stress at that point, respectively; u and v represent the displacement in the x-direction and y-direction, respectively; and K represents the stress in the x-direction and shear stress at that point, respectively. I K I These represent the Type I and Type II fracture factors, respectively. r and θ are the distance and angle between the crack tip and a random location, respectively. k is the elastic modulus of the material; in plane strain problems, k = 3 - 4v; in plane stress problems... A comprehensive assessment is needed to determine the location and stage of the crack's propagation.

[0043] As a preferred technical solution of the present invention: the fourth step specifically involves embedding intelligent particles integrating piezoresistive sensors in a semi-rigid base structure.

[0044] First, when r = R0, measure the stress value σ at the structural point in contact with the smart particle. A The loss function is derived using the least squares method:

[0045]

[0046] Where, σ C denoted as the stress threshold point size, and n represents the number of sampling points.

[0047] The value of σ0 is obtained by taking the extreme value of the derivative of the loss function; the collected stress value σ Ai The monitoring indicators and judgment scheme for a single intelligent particle are derived: when σ Ai When σ > 0, the smart particle is located within the stress field boundary. At this time, the crack is located within a spherical cubic region with the smart particle as the center and R0 as the radius. Ai When σ < 0, the smart particle is located outside the stress field boundary, and the crack tip is located outside the spherical cubic region; the probability that the smart particle detects a crack of length l (l ≥ R) is approximately:

[0048]

[0049] Where V represents the volume of the semi-rigid base layer.

[0050] As a preferred embodiment of the present invention: the fifth step specifically involves randomly arranging a quantity of [number missing] elements in the semi-rigid base layer of the road surface. The smart particles; ideally, the distance between the smart particles is greater than or equal to 2R, and the effective coverage of the structure by the smart particles is expressed as:

[0051]

[0052] Where V is the volume of the semi-rigid base structure. Let be the volume of the spherical cube within the monitoring range of the i-th smart particle; at this point, the probability P(A) of the smart particle detecting the crack is related to the crack length l and the coverage Ω(A) of the smart particle:

[0053] P2 = f(l, Ω(A))

[0054] When a crack initiates, a smart particle at a certain location detects a change in the mechanical response. When σ A >σ C When the crack appears, it is determined that the crack is located within a spherical cubic region with the intelligent particle as the center and R as the radius. As the crack expands, the intelligent particles at different locations collect different mechanical response changes, effectively tracking the location and expansion stage of the crack in real time.

[0055] Beneficial effects:

[0056] Compared with the prior art, this application has the following advantages:

[0057] (1) By using a cross-scale simulation model of the road surface, the process of effectively studying the micro-damage mechanism at the macro-engineering scale was realized;

[0058] (2) The application of intelligent particle sensor device has the advantages of small size, high precision, rich data monitoring data, no damage to the original road structure, and real-time accurate data acquisition through low power Bluetooth transmission;

[0059] (3) Based on the data collected by the smart particles, a method for crack location and crack propagation stage is proposed, realizing an effective monitoring and early warning scheme for cracks in semi-rigid base layers, and providing guidance for subsequent maintenance decisions. Attached Figure Description

[0060] Figure 1 This is a flowchart of the method of the present invention;

[0061] Figure 2 The road structure coupling model for semi-rigid base asphalt pavement provided by this invention;

[0062] Figure 3 This is a schematic diagram showing the relative position of the crack tip and mechanical information of a certain point based on fracture mechanics theory, according to the present invention.

[0063] Figure 4 The layout scheme for monitoring cracks in semi-rigid base courses of asphalt pavement based on intelligent particles provided by this invention;

[0064] Figure 5The schematic diagram of the smart particle integrated with the resistive sensor provided by the present invention. Detailed Implementation

[0065] The invention will now be further described with reference to the accompanying drawings.

[0066] In order to study the crack propagation mechanism of semi-rigid base courses, this invention first uses the finite element-discrete element coupling method to perform numerical simulation calculations to study the crack initiation and propagation process of semi-rigid base courses, thereby deriving a numerical calculation method for crack tracking and location.

[0067] This invention discloses a method for calibrating macroscopic and mesoscopic parameters in a numerical simulation experiment based on finite element and discrete element uniaxial compression. Specifically, a 30mm cube model with a 30*30*30 finite element mesh and 30*30*30 centripetally arranged discrete element particles is established, and a constitutive relation for a linear elastic model is set, while assigning the same density ρ = 2500 kg / m³. 3 The elastic modulus E = 105.20 GPa and Poisson's ratio ν = 0.314 are used to control the same uniaxial compression conditions and properties, and the same number of meshes and particles are used to achieve microscopic matching.

[0068] The macroscopic mechanical response characteristics of the two models were studied by varying the loading mode and the stress-strain control mode, respectively.

[0069] Example 1

[0070] A method for monitoring cracks in semi-rigid base courses of asphalt pavements based on intelligent sensing includes the following steps:

[0071] Step 1: Through matching uniaxial tension and compression numerical simulation experiments, the transformation relationship between macroscopic and microscopic parameters is obtained;

[0072] Step 2: Establish a multi-scale semi-rigid base pavement model using the finite element-discrete element coupling method;

[0073] Step 3: Through numerical simulation of the evolution of cracks in semi-rigid base layers, combined with the Westergard complex variable function method for fracture mechanics type I-II composite cracks, the propagation of cracks in composite semi-rigid base layers is tracked and located.

[0074] Step 4: Smart particles are embedded in the semi-rigid base of the road model to collect information. Based on the response characteristics of stress concentration in the vicinity of cracks in the semi-rigid base, the monitoring indicators of smart particles and the sensitive range are matched and extracted during the crack propagation process.

[0075] Step 5: Establish an intelligent particle layout density scheme for monitoring and locating cracks in semi-rigid base layers, so as to realize the monitoring of the location and expansion stage of cracks in semi-rigid base layers.

[0076] Specifically, the method also includes the following steps:

[0077] Step 1: Fix the bottom surface of the model, and leave the sides unrestricted. Apply a single stress control mode of 1 MPa and a single strain control mode of 1% on the top surface, with a period of 10 Hz and a magnitude of 1 MPa. Then apply a strain control mode of 1% on the top surface and extract the stress and strain state of the model as a whole.

[0078] Step 2: Establish identical cubic models with 30*30*30 finite element meshes and 30*30*30 centripetally arranged discrete element particles. Set the constitutive relation for the linear elastic model, and assign the same density ρ, elastic modulus E, and Poisson's ratio ν to control the same uniaxial tension and compression conditions and properties. Use the same number of meshes and particles to achieve macro-microscopic comparison. Next, change the loading mode. When the model reaches the same state, the correspondence between the microscopic parameters of the particles in the discrete element and the macroscopic parameters of the elements in the finite element is obtained by matching the following formula:

[0079]

[0080]

[0081]

[0082]

[0083]

[0084] Among them, K n For the normal stiffness between particles, K s For shear stiffness, X b For the fracture displacement, F s0 For shear resistance, μ p μ is the coefficient of friction between particles. i C is the inherent coefficient of friction. u T represents uniaxial compressive strength. u d represents tensile strength, and d represents particle diameter.

[0085] Step 3: Extract the displacement u of each element in the finite element model under load. xj u yj u zj , element stress σ xj σ yj σ zj Extract the particle contact force, displacement, and rotation angle from the discrete element model, and calculate the displacement of each particle as u′. xj =X nxi +X sxi 、u′ yj =Xnyi +X syi ,

[0086] u′ zj =X nzi +X szi , where X nxi X sxi X nyi X syi X nzi X szi Let x, y, and z be the x, y, and z components of the relative displacement between the normal and tangential directions of the i-th particle, and let σ′ be the particle contact stress. xj =(F nxi +f sxi ) / S x σ′ yj =(F nyi +F syi ) / S y σ′ zj =(F nzi +F szi ) / S z F nxi F sxi F nyi F syi F nzi F szi These represent the x, y, and z components of the normal and tangential contact forces of the particles, respectively; S x S y S z These represent the contact areas between particles in the x, y, and z directions, respectively.

[0087] Let the strain u be obtained by fitting the particle model. ′ i Stress σ ′ i respectively satisfy f(u) ′ i )=c0+c1i+c2i 2 g(σ) ′ i )=d0+d1i+d2i 2 ,set up Where c0, c1, c2, d0, d1, and d2 are constants, calculate the inner product and substitute it into the normal equation system:

[0088]

[0089]

[0090] Solve for f(u) ′ i)=c0+c1i+c2i 2 g(σ) ′ i )=d0+d1i+d2i 2 Fit u ′ i and σ ′ i Curve; then calculate the standardized Euclidean distance using the following formula:

[0091]

[0092]

[0093] Among them, s, s ′ The standard deviation of particle displacement and stress data; finally, the displacement values ​​at each point in the finite element method are corrected to u. ′ =u±d(u), stress value corrected to σ ′ =σ±d(σ), then the correction value of the elastic modulus of the finite element model. Poisson's ratio correction value v ′ =v±2d(u), the macro-micro transformation relationship is corrected as follows:

[0094]

[0095]

[0096]

[0097]

[0098]

[0099] The present invention also discloses a method for establishing a cross-scale pavement model based on finite element-discrete element coupling, which uses the finite element-discrete element coupling method to establish a cross-scale semi-rigid base pavement model.

[0100] A road structure with an overall size of 8*8*8m was established, with each surface layer being 4+6+8cm and the semi-rigid base layer being 60cm. The surface layer and the semi-rigid base layer were set as finite element elastic element mesh models, and the foundation was set as a finite element Mohr-Coulomb mesh model.

[0101] A discrete element region of 60*60*60cm was selected in the middle of the semi-rigid base, and filled with particles of different sizes to simulate heterogeneous aggregates. 1mm particles were filled to simulate cement binder. Based on the transformation relationship between macroscopic and microscopic material parameters in the above steps, the material parameters of the finite element region and discrete element region of the semi-rigid base of the pavement model were calibrated.

[0102] Stress concentration exists within a certain range in the neighborhood of structural cracks. In the numerical model, particles subjected to significant contact forces are considered elements within the concentrated stress field. The stress field range exhibiting stress concentration in the crack neighborhood is approximated as an irregular column stretched from a spherical cubic region along the crack propagation direction. In the pavement numerical model, by changing the load conditions, after cracks appear in the semi-rigid base layer, the mechanical response of the semi-rigid base layer in the neighborhood of the crack tip at the time of damage is extracted, yielding the relationship between the column and the radius R of the sphere at each stage and direction. i Stress threshold point size σ Ci The average values ​​are used to obtain R0 and σ. C

[0103] According to the theory of linear elastic fracture mechanics, stress concentration exists within a certain range in the neighborhood of structural cracks. Using the Westergard complex variable function method, the expressions for the stress and displacement fields within a certain range in the neighborhood of the tip of a type I-II composite crack are obtained as follows:

[0104]

[0105]

[0106] Where, σ x σ y τ xy These represent the stress in the x-direction, stress in the y-direction, and shear stress at that point, respectively; u and v represent the displacement in the x-direction and y-direction, respectively; and K represents the stress in the x-direction and shear stress at that point, respectively. I K i These represent the Type I and Type II fracture factors, respectively. r and θ are the distance and angle between the crack tip and a random location, respectively. k is the elastic modulus of the material; in plane strain problems, k = 3 - 4v; in plane stress problems... A comprehensive assessment is needed to determine the location and stage of the crack's propagation.

[0107] If the shear modulus G and Poisson's ratio v of the material are known, and the material is assumed to be elastic, then by extracting the mechanical response at random locations in multiple directions, the parameters r, θ, and K can be listed. I K II The four-element equation is used to obtain the values ​​of r and θ, and the location of the crack tip is obtained. This corresponds to the numerical simulation results of the road surface, and the stage of crack propagation is comprehensively judged.

[0108] The method disclosed in this invention for matching and determining the sensitivity range of monitoring indicators acquired during crack propagation based on a single intelligent particle includes the following steps:

[0109] The smart particles, which integrate piezoresistive sensors, are embedded in a semi-rigid base structure. The principle of the smart particles, which integrate piezoresistive strain gauges, is that the material of the sensing element undergoes mechanical deformation under the action of external force, and its resistance value R changes accordingly, which in turn causes the current signal to change, reflecting the mechanical strain. The stress value of the structural point in contact with the smart particles can be measured.

[0110] like Figure 5 As shown

[0111] Depend on have to

[0112] When r = R0, the stress value σ at the structural point in contact with the smart particle is measured. A The loss function is derived using the least squares method:

[0113]

[0114] Where, σ C denoted as the stress threshold point size, and n represents the number of sampling points.

[0115] The value of σ0 is obtained by taking the extreme value of the derivative of the loss function; the collected stress value σ Ai The monitoring indicators and judgment scheme for a single intelligent particle are derived: when σ Ai When σ > 0, the smart particle is located within the stress field boundary. At this time, the crack is located within a spherical cubic region with the smart particle as the center and R0 as the radius. Ai When σ < 0, the smart particle is located outside the stress field boundary, and the crack tip is located outside the spherical cubic region; the probability that the smart particle detects a crack of length l (l ≥ R) is approximately:

[0116]

[0117] Where V represents the volume of the semi-rigid base layer.

[0118] This invention discloses a smart particle placement density and scheme for monitoring and locating cracks in semi-rigid base courses. Specifically, based on the aforementioned individual smart particle monitoring indicators and scheme, a certain number of particles are randomly placed in the semi-rigid base course of the road surface. The smart particles. Ideally, the distance between all smart particles should be greater than or equal to 2R. Then, the effective coverage of the structure by the smart particles is expressed as:

[0119]

[0120] Where V is the volume of the semi-rigid base structure. Let be the volume of the spherical cube within the monitoring range of the i-th smart particle. At this point, the probability P(A) of the smart particle detecting the crack is related to the crack length l and the coverage Ω(A) of the smart particle:

[0121] P(A) = f(l, Ω(A))

[0122] When the crack length l > 2R, the probability of the crack being detected by the smart particles can reach over 70%. If the crack length l > 3R, the probability of the crack being detected reaches 85%. If the crack length l > 5R, the probability of being detected exceeds 95%. By deploying smart particles to collect response information, the location of cracks and the determination of crack propagation stages can be effectively realized.

Claims

1. A method for monitoring cracks in semi-rigid base courses of asphalt pavements based on intelligent sensing, characterized in that, Includes the following steps: Step 1: Perform parameter matching through uniaxial compression simulation experiments to obtain the macro-micro parameter transformation relationship of the model; Step 2: Establish a multi-scale semi-rigid base pavement model using the finite element-discrete element coupling method; Step 3: Through numerical simulation of the evolution of cracks in semi-rigid base layers, combined with the Westergard complex variable function method for fracture mechanics type I-II composite cracks, the propagation of cracks in composite semi-rigid base layers is tracked and located. Step 4: Smart particles are embedded in the semi-rigid base of the road model to collect information. Based on the response characteristics of stress concentration in the vicinity of cracks in the semi-rigid base, the monitoring indicators of smart particles and the sensitive range are matched and extracted during the crack propagation process. Step 5: Establish an intelligent particle layout density scheme for monitoring and locating cracks in semi-rigid base layers, so as to realize the monitoring of the location and expansion stage of cracks in semi-rigid base layers.

2. The method for monitoring cracks in semi-rigid base courses of asphalt pavement based on intelligent sensing according to claim 1, characterized in that... The specific steps of the first step are as follows: First, establish a cubic model of the same size with a 30*30*30 finite element mesh and 30*30*30 centripetally arranged discrete element particles, respectively. Set the constitutive relation of the linear elastic model and assign the same density ρ = 2500 kg / m³. 3 The model was designed with an elastic modulus E = 105.20 GPa and a Poisson's ratio ν = 0.314 to control the same uniaxial tension and compression conditions and properties. The same number of meshes and particles were used to achieve macro-microscopic comparison. Next, the bottom surface of the model was fixed, while the sides were unrestricted. A single stress control mode of 1 MPa and a single strain control mode of 1% were applied to the top surface. A stress mode with a period of 10 Hz and a magnitude of 1 MPa, and a strain mode with a period of 10 Hz and a magnitude of 1% were also applied. The overall stress and strain state of the model was extracted. The correspondence between the microscopic parameters of the particles in the discrete element method and the macroscopic parameters of the elements in the finite element method was obtained by matching the following formula: Among them, K n For the normal stiffness between particles, K s For shear stiffness, X b For the fracture displacement, F s0 For shear resistance, μ p μ is the coefficient of friction between particles. i C is the inherent coefficient of friction. u T represents uniaxial compressive strength. u Where d is the tensile strength, d is the particle diameter, and i is the number of any element; Next, extract the displacement u of each element in the finite element model in the x, y, and z directions under load. xj u yj u zj and the stress σ in the x, y, and z directions. xj σ yj σ zj Extract the particle contact force, displacement, and rotation angle from the discrete element model, and calculate the displacement of each particle as u'. xj =X nxi +X sxi 、u' yj =X nyi +X syi , u' zj = X nzi + X szi , where X nxi , X sxi , X nyi , X syi , X nzi , X szi Let x, y, and z be the x, y, and z components of the relative displacement between the normal and tangential directions of the i-th particle, and let σ' be the particle contact stress. xj =(F nxi +F sxi ) / S x ,σ' yj =(F nyi +F syi ) / S y ,σ' zj =(F nzi +F szi ) / S z F nxi F sxi F nyi F syi F nzi F szi These represent the x, y, and z components of the normal and tangential contact forces of the particles, respectively; S x S y S z These represent the contact areas between particles in the x, y, and z directions, respectively. Let the strain u' be fitted by the particle model. i Stress σ' i respectively satisfy f(u') i )=c0+c1i+c2i 2 g(σ') i )=d0+d1i+d2i 2 ,set up Where c0, c1, c2, d0, d1, and d2 are constants, calculate the inner product and substitute it into the normal equation system: Solve for f(u') i )=c0+c1i+c2i 2 g(σ') i )=d0+d1i+d2i 2 Fit u' i and σ' i Curve; then calculate the standardized Euclidean distance using the following formula: Among them, u j σ represents the strain fitted by the j-th particle model. j Let s represent the stress fitted by the model of the j-th particle. j The standard deviation of particle displacement data, s' j The standard deviation of the particle stress data is represented by u'. Finally, the displacement values ​​at each point in the finite element method are corrected to u'. j =u j ±d(u), stress value corrected to σ' j =σ j ±d(σ), then the elastic modulus correction value of the finite element model The Poisson ratio correction value v' = v ± 2d(u) is used to correct the macro-micro parameter transformation relationship as follows:

3. The method for monitoring cracks in semi-rigid base courses of asphalt pavement based on intelligent sensing according to claim 2, characterized in that... The second step involves establishing a pavement structure with an overall size of 8*8*8m, ​​each surface layer being 4+6+8cm and the semi-rigid base layer being 60cm. The surface layer and the semi-rigid base layer are set as finite element elastic element mesh models, and the foundation is set as a finite element Mohr-Coulomb mesh model. A discrete element region of 60*60*60cm is selected in the middle of the semi-rigid base layer and filled with particles of different sizes to simulate heterogeneous aggregates. 1mm particles are filled to simulate cement binder. Based on the modified macro-micro parameter conversion relationship described in claim 2, the material parameters of the finite element region and the discrete element region of the semi-rigid base layer are calibrated.

4. The method for monitoring cracks in semi-rigid base courses of asphalt pavement based on intelligent sensing according to claim 1, characterized in that... The third step involves approximating the stress field range where stress concentration occurs in the crack neighborhood as an irregular column stretched from a spherical cubic region along the crack propagation direction; changing the load conditions in the pavement numerical model, and extracting the mechanical response of the semi-rigid base layer near the crack tip when cracks appear, to obtain the radius R of the sphere at each stage. i Stress threshold point size σ Ci The average value of the radius of the stress field sphere, R0, and the average value of the stress threshold, σ, are obtained by averaging. C ; Based on the Westergard complex variable function method for I-II composite cracks in fracture mechanics, the stress and displacement values ​​at the above points are used to calculate r and θ, and the location and propagation stage of the crack are determined comprehensively. Where, σ x σ y τ xy Let u and v represent the stress in the x-direction, stress in the y-direction, and shear stress at this point, respectively; and let v and v represent the displacement in the x-direction and y-direction, respectively. K I K II These represent the Type I and Type II fracture factors, respectively. r and θ are the distance and angle between the crack tip and a random location, respectively. k is the elastic modulus of the material; in plane strain problems, k = 3 - 4v; in plane stress problems... G represents the shear modulus of the finite element method.

5. A method for monitoring cracks in semi-rigid base courses of asphalt pavement based on intelligent sensing, as described in claim 1, characterized in that... The fourth step involves embedding a smart particle integrated with a piezoresistive sensor in the semi-rigid base structure, and measuring the stress value σ at the structural point in contact with the smart particle when the distance r between the crack tip and the smart particle is equal to the average value R0 of the spherical radius of the stress field. A The loss function is derived using the least squares method: Where, σ C The stress threshold point size is given by n, which represents the number of sampling points. The stress judgment threshold σ0 is determined by differentiating the loss function and taking its extreme value; the stress judgment threshold σ0 is then compared with the stress value σ collected by the i-th smart particle. Ai By comparison, the monitoring indicators and judgment schemes for a single intelligent particle are derived: when σ Ai When σ > 0, the smart particle is located within the stress field boundary. At this time, the crack is located within a spherical cubic region with the smart particle as the center and R0 as the radius. Ai When σ < 0, the smart particle is located outside the stress field boundary, and the crack tip is located outside the spherical cubic region; the probability that the smart particle detects a crack of length l is approximately: Where V represents the volume of the semi-rigid base structure.

6. The method for monitoring cracks in semi-rigid base courses of asphalt pavement based on intelligent sensing according to claim 5, characterized in that, The fifth step specifically involves randomly arranging a quantity of [number missing] elements in the semi-rigid base layer of the road surface. The smart particles; ideally, the distance between the smart particles is greater than or equal to 2R0, and the effective coverage of the structure by the smart particles is expressed as: Where V is the volume of the semi-rigid base structure, V A The volume range that the intelligent particle array can monitor; Let be the volume of the spherical cube within the monitoring range of the nth smart particle; at this point, the probability P(A) of the smart particle detecting the crack is related to the crack length l and the coverage Ω(A) of the smart particle: P(A)=f(l,Ω(A)) When a crack initiates, a smart particle at a certain location detects a change in the mechanical response. When σ A >σ C When the crack appears, it is determined that the crack is located within a spherical cubic region with the intelligent particle as the center and R as the radius. As the crack expands, the intelligent particles at different locations collect different mechanical response changes, effectively tracking the location and expansion stage of the crack in real time.

Images