An asphalt pavement fatigue life prediction method for automatic driving platoon

Abstract

The application discloses an asphalt pavement fatigue life prediction method for automatic driving platoon, and comprises the following steps: firstly, finite element modeling is carried out on a typical asphalt pavement structure, and numerical simulation of an environmental temperature field is realized; then, a tire grounding shape is simplified into a circular double-wheel load, and a moving load is applied according to the axle load characteristics of a typical truck; then, combined working condition simulation is carried out according to the speed, inter-vehicle distance and other characteristics of the automatic driving platoon, and a maximum longitudinal tensile strain waveform diagram of an asphalt layer bottom is obtained; mechanical response waveform characteristics are analyzed, and independent shape parameters are obtained through a main level analysis method; equivalent damage under automatic driving platoon driving is calculated based on the Miner linear cumulative damage theory; and an asphalt pavement fatigue life model based on waveform characteristics is established through a numerical fitting method according to the equivalent damage principle. The application provides a method for road structure design under an automatic driving scene.

Description

Technical Field

[0001] This invention relates to a method for predicting the fatigue life of asphalt pavement, and more particularly to a method for predicting the fatigue life of asphalt pavement for autonomous driving formations. Background Technology

[0002] With the rapid development and deep integration of next-generation information technologies such as vehicle-to-everything (V2X) and artificial intelligence, key technologies for autonomous driving are maturing, and application scenarios are becoming increasingly clear. In platooning, the periodicity of the load changes significantly and exhibits different characteristics depending on the platooning configuration. This change in load pattern caused by driving and platooning modes profoundly affects the stress state of the pavement structure and thus its performance. Furthermore, current methods for predicting the fatigue life of asphalt pavement structures are only applicable to typical manned driving scenarios and do not consider the characteristics of autonomous platooning. Therefore, methods for predicting the fatigue life of asphalt pavement structures for autonomous truck platooning urgently need optimization. Summary of the Invention

[0003] Purpose of the invention: The purpose of this invention is to provide a method for predicting the fatigue life of asphalt pavement for autonomous driving formations. This method involves three-dimensional modeling of typical asphalt pavement structures, proposing independent characteristic parameters that characterize the maximum tensile strain at the bottom of the asphalt layer, and changing the focus dimension from the traditional number of axle loads to the number of platooning trips. By applying the equivalent damage principle, a fatigue life prediction formula for asphalt pavement structures is obtained.

[0004] Technical solution: The method for predicting the fatigue life of asphalt pavement of the present invention includes the following steps:

[0005] S1. Based on actual engineering problems, obtain the material parameters of typical asphalt pavement structures and establish a three-dimensional finite element model of asphalt pavement.

[0006] S2 simplifies the tire contact shape into a circular dual-wheel load, and writes the Dload subroutine based on the axle load characteristics of a typical truck to apply the periodic moving load on the three-dimensional asphalt pavement finite element model.

[0007] S3, based on the characteristics of autonomous driving formation, combines the features of vehicle speed, number of vehicles, left and right vehicle spacing, and front and rear vehicle spacing under different working conditions and simulates them to obtain the waveform diagram of the maximum longitudinal tensile strain at the bottom of the asphalt layer under each working condition.

[0008] S4. Analyze the waveform characteristics of the maximum longitudinal tensile strain at the bottom of the asphalt layer under all working conditions, and obtain the main independent shape parameters that can characterize this type of waveform characteristics through the principal hierarchical analysis method.

[0009] S5, based on Miner's linear cumulative damage principle, calculates the cumulative fatigue damage under each working condition;

[0010] S6, based on the principle of equivalent damage, establishes a fatigue life model of asphalt pavement based on waveform characteristics through a multivariate nonlinear fitting method.

[0011] Furthermore, in step S2, the specific steps for applying the periodic moving load on the three-dimensional asphalt pavement finite element model are as follows:

[0012] S21 simplifies the tire contact shape to a double circular load, assuming that the pressure is evenly distributed on the contact surface;

[0013] S22, select a typical truck to analyze the axle load distribution, including the number of axles, wheelbase, and axle load. Combined with double circular uniformly distributed load, use the Dload subroutine to realize the load effect of a single truck.

[0014] S23, based on the truck's body length l, the distance between vehicles in the autonomous driving platoon x, the vehicle speed v, and the number of vehicles n, calculate the cycle time t:

[0015] t=((l*n+x*(n-1)) / v

[0016] In the Dload subroutine, the MOD function is used to set STEPTIME to achieve the cyclic movement of loads in the formation.

[0017] Furthermore, in step S3, the specific implementation steps for obtaining the waveform diagram of the maximum longitudinal tensile strain at the bottom of the asphalt layer under each working condition are as follows:

[0018] S31, for different characteristics of autonomous driving platooning, determines the range of variation of each variable and combines each variable into different working conditions;

[0019] S32, writes corresponding ".FOR" files for different working conditions, applies the Dload subroutine to the established three-dimensional asphalt pavement finite element model, and outputs the waveform of the maximum longitudinal tensile strain at the bottom of the asphalt layer under each working condition.

[0020] Furthermore, in step S4, the detailed implementation steps for obtaining the main independent shape parameters that characterize this type of waveform using the principal hierarchical analysis method are as follows:

[0021] S41, based on the waveform diagram of the maximum longitudinal tensile strain at the bottom of the asphalt layer under all working conditions, proposes m graphical parameters X1, X2, ..., X from the aspects of loading signal, vehicle speed, and pavement structure to characterize this type of waveform. m ;

[0022] S42, establish the m-dimensional graphical parameters of n working conditions as a matrix X = [X1, X2, X3, ..., X...]. m ], where the set of working conditions corresponding to the j-th dimension is 1≤j≤n, r is the number of working conditions; normalize matrix X to obtain matrix

[0023] S43, Calculate the matrix The covariance matrix C;

[0024] S44, calculate the eigenvalues ​​and eigenvectors of the covariance matrix C, sort the eigenvalues ​​from largest to smallest, and retain the first N eigenvalues. The eigenvectors corresponding to the first N eigenvalues ​​are X. j Principal components P1, P2, ..., P N ;

[0025] S45, Select principal components P1, P2, ..., P N Graphical parameters X1, X2, ..., X with high similarity N , as an independent shape parameter characterizing the waveform features of the maximum longitudinal tensile strain at the bottom of the asphalt layer under autonomous driving platooning.

[0026] Furthermore, in step S5, the detailed steps for calculating the cumulative fatigue damage under each working condition are as follows:

[0027] S51, based on the fatigue life calculation formula, calculate the fatigue life N corresponding to the maximum longitudinal tensile strain at the bottom of the asphalt layer at each time step under each working condition. f1 :

[0028]

[0029] in,

[0030] N f1 — Fatigue cracking life of asphalt mixture layers;

[0031] β—Target reliability index;

[0032] k a —Adjustment coefficient for seasonally frozen soil regions;

[0033] k b — Fatigue mode loading coefficient;

[0034] E a —Dynamic compression modulus of asphalt mixture at 20℃, unit: MPa;

[0035] VFA – Asphalt saturation of asphalt mixture, %;

[0036] k T1 —Temperature adjustment coefficient;

[0037] ε a —Tensile strain at the bottom of each asphalt mixture layer, 10 -6 ;

[0038] S52, based on Miner's linear damage accumulation theory, calculate the cumulative fatigue damage D during autonomous driving platooning:

[0039]

[0040] Here, step represents the number of time steps under each working condition.

[0041] Furthermore, in step S6, the detailed implementation steps for establishing an asphalt pavement fatigue life model based on waveform features using a multivariate nonlinear fitting method are as follows:

[0042] S61, Based on the equivalent damage theory, a fatigue life prediction formula model for asphalt pavement based on independent shape parameters of mechanical response waveform is established for autonomous driving formation:

[0043] F(N f0 ) = f(X1, X2, ..., X N )

[0044] F(X k ) = f(n, v, x, y)

[0045] N f0 — Fatigue cracking life of asphalt mixture layers;

[0046] X k —Independent graphical parameters of the maximum longitudinal tensile strain waveform at the bottom of the asphalt layer, k = 1, 2, ..., N;

[0047] n—Number of vehicles;

[0048] v — Speed ​​of autonomous driving platooning vehicles, km / h;

[0049] x — Spacing between vehicles in autonomous driving platooning, in meters;

[0050] y — Spacing between vehicles in autonomous driving platoons, in meters;

[0051] S62, by using simulation software to perform nonlinear fitting on the fatigue life prediction formula model and independent graphical parameters, a fatigue life prediction formula for asphalt pavement oriented towards autonomous driving formations is obtained.

[0052] Compared with the prior art, the significant advantages of this invention are as follows:

[0053] This invention addresses future autonomous driving platooning scenarios, considering numerous differences between autonomous driving platooning and manned driving scenarios, such as periodic loads due to algorithm controllability, significant differences in vehicle speed and spacing, and the fixed nature of wheel tracks within lanes. It performs three-dimensional modeling of typical asphalt pavement structures and proposes independent characteristic parameters representing the maximum tensile strain at the bottom of asphalt layers. Based on this, it shifts the focus from the traditional number of axle loads to the number of platooning trips. Through the principle of equivalent damage, it further fits and derives a novel formula for predicting the fatigue life of asphalt pavement structures, providing new insights for road structure design methods in autonomous driving scenarios. Attached Figure Description

[0054] Figure 1 This is a flowchart of the present invention;

[0055] Figure 2 This is a schematic diagram of the double-circle uniformly distributed load in this invention;

[0056] Figure 3 This is a waveform diagram showing the maximum longitudinal tensile strain at the bottom of the asphalt layer in this invention. Detailed Implementation

[0057] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0058] This invention provides a method for calculating the fatigue life of asphalt pavement structures under repeated platooning loads, assuming autonomous driving formations operate at specific formations and speeds. The following specific example further illustrates this method for predicting the fatigue life of asphalt pavement, and its process is described in... Figure 1 As shown.

[0059] Step 1: In the Part module, establish a three-dimensional large-scale asphalt pavement finite element model, use symmetrical load boundary conditions to reduce the requirements on the computer, divide the pavement into layers, and assign corresponding material properties to each structural layer in the Property module.

[0060] Step 2: Simplify the tire contact shape into a circular dual-wheel load, and write the Dload subroutine based on the axle load characteristics of a typical truck to apply the periodic moving load on the three-dimensional asphalt pavement finite element model.

[0061] Step 21: Simplify the tire contact patch shape to a double-circle load, assuming the pressure is uniformly distributed on the contact surface. The simplified model is as follows: Figure 2 As shown. The equivalent circle diameter is d = 0.213 m, the center-to-center distance between the two circles is 1.5 d, and the tire contact pressure is 0.7 MPa.

[0062] Step 22: Select a typical truck to analyze axle load distribution. The wheelbase is 2000+4500+1350mm, the front and rear track widths are 2050 / 1878mm, the front and rear overhangs are 1.515 / 2.635m, and the truck width is 2.55m. Combined with a double-circle uniformly distributed load, the Dload subroutine is used to implement the load on a single truck.

[0063] Step 23, when the vehicle length l = 12m, the distance between vehicles in the autonomous driving platoon x = 5m, the vehicle speed v = 80km / h, and the number of vehicles n = 4, calculate the cycle time t:

[0064]

[0065] In the Dload subroutine, the MOD function is used to set STEPTIME to achieve the cyclic movement of loads in the formation.

[0066] Step 3: Based on the characteristics of autonomous driving formation, combine features such as vehicle speed, number of vehicles, left and right vehicle spacing, and front and rear vehicle spacing under different working conditions and simulate them to obtain the waveform diagram of the maximum longitudinal tensile strain at the bottom of the asphalt layer under each working condition.

[0067] Step 31: Based on the different characteristics of autonomous driving platooning, the vehicle speed varies from 60km / h to 120km / h, the number of vehicles varies from 2 vehicles / column to 4 vehicles / column, the distance between vehicles in front and behind varies from 3.5m to 5.5m, and the distance between vehicles on the left and right varies from 0.6m to 1.4m. The above variables are combined into 225 different working conditions.

[0068] Step 32: Create corresponding ".FOR" files for different working conditions, apply the Dload subroutine to the established three-dimensional asphalt pavement finite element model, and output the waveform of the maximum longitudinal tensile strain at the bottom of the asphalt layer under each working condition. Figure 3 The waveform diagram of the maximum longitudinal tensile strain at the bottom of the asphalt layer under the condition of a convoy of 4 vehicles in a single line, a speed of 80km / h, a front-to-back distance of 5.0m, and a left-to-right distance of 1.0m.

[0069] Step 4: Analyze the waveform characteristics of the maximum longitudinal tensile strain at the bottom of the asphalt layer under all working conditions, and obtain the main independent shape parameters that can characterize this type of waveform characteristics through the principal hierarchical analysis method.

[0070] Step 41: Based on the waveform diagram of the maximum longitudinal tensile strain at the bottom of the asphalt layer under all working conditions, propose 10 graphical parameters X1, X2, ..., X6 to characterize this type of waveform from aspects such as loading signal, vehicle speed, and pavement structure. 10 For example, X5 represents the peak value of the maximum tensile strain ε at the bottom of the asphalt layer during a single unmanned platooning operation.

[0071] Step 42: Establish the 10 graphical parameters of the above 225 working conditions into a matrix X = [X1, X2, X3, ..., X...]. 10 ], where the set of working conditions corresponding to the j-th dimension is For example:

[0072] X5=(4.87E-05, 4.92E-05,..., 3.68E-05, 3.85E-05) T

[0073] Next, normalize matrix X to obtain matrix

[0074] Step 43, calculate the matrix covariance matrix

[0075]

[0076] Step 44: Calculate the eigenvalues ​​and eigenvectors of the covariance matrix C. Sort the eigenvalues ​​from largest to smallest, retaining the first three eigenvalues ​​for better interpretability: 4.81, 3.01, and 1.55. The eigenvectors corresponding to the first three eigenvalues ​​are the dataset's eigenvectors. Principal components P1, P2, P3:

[0077] P1=(0.908, 2.576, 3.897,..., -3.456, -1.160, 0.315) T

[0078] P2=(-2.143,-0.950,0.184,...,-0.674,-0.545,-0.038) T

[0079] P3=(-0.937,-1.521,-2.366,...,1.558,1.234,0.111) T

[0080] Step 45: Select the graphical parameters X1, X2, and X3, which are highly correlated with the principal components P1, P2, and P3, as independent shape parameters to characterize the waveform features of the maximum longitudinal tensile strain at the bottom of the asphalt layer under autonomous driving platooning.

[0081] Step 5: Based on Miner's linear cumulative damage principle, calculate the cumulative fatigue damage under each working condition;

[0082] Step 51: Based on the fatigue life calculation formula in the existing specifications, calculate the fatigue life N corresponding to the maximum longitudinal tensile strain at the bottom of the asphalt layer at each time step under each working condition. f1 .

[0083]

[0084] N f1 — Fatigue cracking life of asphalt mixture layer (axle cycles);

[0085] β—Target reliability index, which is 1.65 in this embodiment;

[0086] k a —The seasonal permafrost region adjustment coefficient is 1.0 in this embodiment;

[0087] k b —The fatigue mode loading coefficient, calculated in this embodiment, is 0.0686;

[0088] E a —The dynamic compression modulus (MPa) of the asphalt mixture at 20°C, which is 10000 in this embodiment;

[0089] VFA—Asphalt saturation (%) of the asphalt mixture, which is 70 in this embodiment;

[0090] k T1 —Temperature adjustment coefficient, which is 1.23 in this embodiment;

[0091] ε a — Tensile strain at the bottom of asphalt mixture layer (10 -6 );

[0092] Step 52: Calculate the cumulative fatigue damage of autonomous driving platooning based on Miner's linear damage accumulation theory.

[0093]

[0094] step — the number of time steps under each working condition;

[0095] The cumulative damage calculated for the 225 working conditions in this embodiment is:

[0096] D=(1.69E-07, 2.66E-06,..., 6.41E-07, 1.03E-06) T

[0097] Step 6: Based on the principle of equivalent damage, establish an asphalt pavement fatigue life model based on waveform characteristics using a multivariate nonlinear fitting method. This includes the following steps:

[0098] Step 61: Based on the equivalent damage theory, establish a formula model for predicting the fatigue life of asphalt pavement based on independent shape parameters of mechanical response waveforms for autonomous driving formations.

[0099]

[0100] F(X i )=f(n, v, x, y), i=1, 2, 3 (5)

[0101] N f0 — Fatigue cracking life of asphalt mixture layer (number of platooning trips);

[0102] X1, X2, X3 — Independent graphical parameters of the maximum longitudinal tensile strain waveform at the bottom of the asphalt layer;

[0103] A, B, C, D, E, F, G — Fitting parameters;

[0104] n—Number of vehicles;

[0105] v — Autonomous driving platooning speed (km / h);

[0106] x — Spacing between vehicles in autonomous driving platoons (m);

[0107] y — Spacing between vehicles in autonomous driving platoons (m);

[0108] Step 62: Using Matlab software, fit the above fatigue life prediction formula model and independent graphical parameters with power functions to obtain parameters A, B, C, D, E, F, and G.

[0109]

[0110] f(X1) = 5.8 * 10 -7 n + 3.5 * 10 -9 v-4.6*10 -7 x-3.4*10 -7 y+2.4*10 -6 (7)

[0111] f(X2) = 0.003n + 7.2 * 10 -5 v + 0.017x - 0.044y + 0.4 (8)

[0112] f(X3)=0.082n-0.002v+0.004x+0.016y+0.2 (9)

[0113] Based on the fitting results, the fatigue life of asphalt pavement structure under autonomous driving platooning conditions can be calculated. For example, when there are 4 autonomous vehicles in a single platoon, with a speed of 80 km / h, a front-to-back distance of 4.5 m, and a left-to-right distance of 1.2 m, the cumulative fatigue life is 548,971 platoon trips, corresponding to a cumulative fatigue damage of 1.82E-06, which is only 5.13% different from the damage calculated by the fatigue life prediction formula in the standard.

[0114] The embodiments are merely illustrative of the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made to the technical solution based on the technical concept proposed in this invention shall fall within the scope of protection of this invention.

Claims

1. A method for predicting the fatigue life of asphalt pavement for autonomous driving formations, characterized in that, Includes the following steps: S1. Based on actual engineering problems, obtain the material parameters of typical asphalt pavement structures and establish a three-dimensional finite element model of asphalt pavement. S2 simplifies the tire contact shape into a circular dual-wheel load, and writes the Dload subroutine based on the axle load characteristics of a typical truck to apply the periodic moving load on the three-dimensional asphalt pavement finite element model. S3, based on the characteristics of autonomous driving formation, combines the features of vehicle speed, number of vehicles, left and right vehicle spacing, and front and rear vehicle spacing under different working conditions and simulates them to obtain the waveform diagram of the maximum longitudinal tensile strain at the bottom of the asphalt layer under each working condition. S4. Analyze the waveform characteristics of the maximum longitudinal tensile strain at the bottom of the asphalt layer under all working conditions, and obtain the main independent shape parameters that can characterize this type of waveform characteristics through the principal hierarchical analysis method. S5, based on Miner's linear cumulative damage principle, calculates the cumulative fatigue damage under each working condition; S6. Based on the principle of equivalent damage, a fatigue life model of asphalt pavement based on waveform characteristics is established by using a multivariate nonlinear fitting method. In step S2, the specific steps for applying the periodic moving load on the three-dimensional asphalt pavement finite element model are as follows: S21 simplifies the tire contact shape to a double circular load, assuming that the pressure is evenly distributed on the contact surface; S22, select a typical truck to analyze the axle load distribution, including the number of axles, wheelbase, and axle load. Combined with double circular uniformly distributed load, use the Dload subroutine to realize the load effect of a single truck. S23, calculate the cycle time based on the truck body length l, the distance between vehicles in the autonomous driving platoon x, the vehicle speed v, and the number of vehicles n. : In the Dload subroutine, the MOD function is used to set STEPTIME to achieve the cyclic movement of loads in the formation.

2. The method for predicting the fatigue life of asphalt pavement for autonomous driving formations according to claim 1, characterized in that, In step S3, the specific steps for obtaining the waveform diagram of the maximum longitudinal tensile strain at the bottom of the asphalt layer under each working condition are as follows: S31, for different characteristics of autonomous driving platooning, determines the range of variation of each variable and combines each variable into different working conditions; S32, writes corresponding ".FOR" files for different working conditions, applies the Dload subroutine to the established three-dimensional asphalt pavement finite element model, and outputs the waveform of the maximum longitudinal tensile strain at the bottom of the asphalt layer under each working condition.

3. The method for predicting the fatigue life of asphalt pavement for autonomous driving formations according to claim 1, characterized in that, In step S4, the detailed implementation steps for obtaining the main independent shape parameters that characterize this type of waveform using the principal hierarchical analysis method are as follows: S41, based on the waveform diagram of the maximum longitudinal tensile strain at the bottom of the asphalt layer under all working conditions, proposes m graphical parameters to characterize this type of waveform from the aspects of loading signal, vehicle speed, and pavement structure. ; S42, establish the m-dimensional graphical parameters of n working conditions as a matrix. The set of working conditions corresponding to the j-th dimension is: , 1≤j≤n, r is the number of working conditions; normalize matrix X to obtain matrix ; S43, Calculate the matrix covariance matrix ; S44, calculate the eigenvalues ​​and eigenvectors of the covariance matrix C, sort the eigenvalues ​​from largest to smallest, and retain the first N eigenvalues. The eigenvectors corresponding to the first N eigenvalues ​​are... principal components ; S45, Select with principal component Highly similar graphic parameters , as an independent shape parameter characterizing the waveform features of the maximum longitudinal tensile strain at the bottom of the asphalt layer under autonomous driving platooning.

4. The method for predicting the fatigue life of asphalt pavement for autonomous driving formations according to claim 1, characterized in that, In step S5, the detailed steps for calculating the cumulative fatigue damage under each working condition are as follows: S51, based on the fatigue life calculation formula, calculate the fatigue life corresponding to the maximum longitudinal tensile strain at the bottom of the asphalt layer at each time step under each working condition. : in, — Fatigue cracking life of asphalt mixture layers; —Target reliability indicators; —Adjustment coefficient for seasonally frozen soil regions; — Fatigue mode loading coefficient; —Dynamic compression modulus of asphalt mixture at 20℃, unit: MPa; —Asphalt saturation of asphalt mixture, % —Temperature adjustment coefficient; —Tensile strain at the bottom of each asphalt mixture layer, 10 -6 ; S52, based on Miner's linear damage accumulation theory, calculate the cumulative fatigue damage D during autonomous driving platooning: Here, step represents the number of time steps under each working condition.

5. The method for predicting the fatigue life of asphalt pavement for autonomous driving formations according to claim 1, characterized in that, In step S6, the detailed implementation steps for establishing the asphalt pavement fatigue life model based on waveform characteristics using a multivariate nonlinear fitting method are as follows: S61, Based on the equivalent damage theory, a fatigue life prediction formula model for asphalt pavement based on independent shape parameters of mechanical response waveform is established for autonomous driving formation: — Fatigue cracking life of asphalt mixture layers; —Independent graphical parameters of the maximum longitudinal tensile strain waveform at the bottom of the asphalt layer, k=1,2,…,N; —Number of vehicles; —Speed ​​of autonomous driving platooning vehicles, km / h; — Spacing between vehicles in autonomous driving platoons, in meters; —Left and right distance between vehicles in autonomous driving platoons, in meters; S62, by using simulation software to perform nonlinear fitting on the fatigue life prediction formula model and independent graphical parameters, a fatigue life prediction formula for asphalt pavement oriented towards autonomous driving formations is obtained.

Images