Multiphase heterogeneous model construction method for forward modeling of three-dimensional ground penetrating radar

Abstract

The invention discloses a multiphase heterogeneous model construction method for three-dimensional ground penetrating radar forward modeling. The method comprises the following steps: investigating a typical structure of a highway asphalt pavement; setting the size and network of the multiphase heterogeneous model; calculating a coarse and rapid material gradation curve of the surface layer and the base layer; reconstructing the form of the coarse aggregate, exporting the coordinate and radius of the small ball, and calculating the volume fraction of the fine aggregate, the pore and the asphalt; creating an hdf5 array of a random phase and a dielectric property file; carrying out secondary reconstruction on the coarse rapid material in gprmax; creating an hdf5 array and a dielectric property file of the coarse aggregate; and establishing a three-dimensional multiphase heterogeneous electromagnetic score checking model. According to the method, the full-profile map characteristics of the highway structure disease echoes under heterogeneous interference can be obtained.

Description

Technical Field

[0001] This invention belongs to the field of geophysical simulation and measurement technology, specifically relating to a method for constructing a multiphase heterogeneous model for three-dimensional ground-penetrating radar forward modeling. Background Technology

[0002] Due to the complex internal environment of highway structures, the reflection patterns of electromagnetic waves in different underground media are unclear. Therefore, detecting structural defects in highways requires prior forward modeling to clarify the waveform characteristics and propagation patterns of different defect types on radar maps, providing prior knowledge for identifying abnormal areas in radar maps. Existing studies often treat pavement structural materials as equivalent to layered homogeneous media, assuming that the materials constituting each structural layer are uniformly distributed and using equivalent dielectric constants to characterize the dielectric constant of the entire structural layer. However, real pavement materials are multiphase non-homogeneous mixtures composed of aggregate particles, asphalt mortar, cement mortar, and internal pores, exhibiting typical non-homogeneous characteristics such as varying volume contents, different geometric shapes, random spatial locations, and different dielectric properties. Homogeneous models cannot reflect the differences in electromagnetic properties of non-homogeneous pavement materials at different locations, nor can they characterize the degree of attenuation of electromagnetic waves due to non-homogeneous random disturbances within the structure. This leads to overly idealized forward modeling results that differ significantly from actual field measurements. To circumvent the aforementioned problems, existing research has developed a random dielectric model to characterize the dielectric constant distribution of each phase material. This involves using a dielectric constant tester to scan the dielectric constant distribution of each phase material on a cross-section, and then inputting the obtained distribution characteristics into an electromagnetic differential model to characterize the non-uniform properties of the material. However, many problems still exist, such as the fact that the dielectric constant is a two-dimensional random distribution, which cannot characterize the non-uniform distribution of road surface materials in three-dimensional space, ignoring the anisotropy of aggregate shapes with typical morphological characteristics, and making it difficult to reflect the true path of electromagnetic wave propagation.

[0003] Therefore, there is an urgent need for a ground-penetrating radar forward modeling method that comprehensively measures the heterogeneous characteristics of road surface materials in order to improve the realism and accuracy of forward modeling maps. Summary of the Invention

[0004] The purpose of this invention is to address the above problems by proposing a method for constructing a multiphase heterogeneous model for three-dimensional ground-penetrating radar forward modeling. Based on discrete element method, the method reconstructs the coarse aggregate morphology of each layer of asphalt pavement and realizes the random distribution of dielectric properties of asphalt, fine aggregate, and pore three-phase media based on random medium theory, which is beneficial to improving the accuracy of ground-penetrating radar forward modeling.

[0005] To address at least one of the aforementioned technical problems, according to one aspect of the present invention, a method for constructing a multiphase heterogeneous model for three-dimensional ground-penetrating radar forward modeling is provided, comprising the following steps: Based on the typical structural composition of highway asphalt pavement, the surface layer consists of S layers, each layer is paved with MS1, MS2, and MS3, and the thickness of each layer is HS1, HS2, and HS3. The base layer consists of B layers, each layer is paved with BS1 and BS2, and the thickness of each layer is HS1 and HS2. Based on the typical structural composition of highway asphalt pavement, the model size is determined to be X. Y Z, based on the current accuracy of 3D ground-penetrating radar detection, sets the model mesh size to dx. dy dz; Calculate the gradation curves of MS1, MS2, and MS3 for each layer of asphalt pavement materials. The asphalt mixture is divided into coarse and fine aggregates with a particle size of 2.36mm. Based on the gradation ratio of coarse and fine aggregates, the gradation curve of coarse aggregates for each layer of pavement materials in the pavement layer is determined. Calculate the gradation curves of BS1 and BS2 for each layer of base course materials. The cement mixture is divided into coarse and fine aggregates with a particle size of 4.75mm. Based on the gradation ratio of coarse and fine aggregates, the gradation curve of coarse aggregates for each layer of base course materials is determined. Based on the gradation curves of coarse aggregates in each layer, a coarse aggregate reconstruction model is generated in PFC 3D software, and the coordinates and radii of the small spheres that make up the coarse aggregates are exported. The porosity of each layer in the surface layer is set to ΦS1, ΦS2, and ΦS3, and the asphalt-aggregate ratio of each layer is set to δS1, δS2, and δS3.

[0006] Assuming a constant aggregate mass in the asphalt mixture, calculate the volume fractions of the fine aggregate phase, porous phase, and asphalt phase; set the porosity of each layer of the base course to ΦB1 and ΦB2, and the water-cement ratio of each layer to ηB1 and ηB2. Assuming a constant aggregate mass in the cement mixture, calculate the volume fractions of the fine aggregate phase, porous phase, and cement phase. In MATLAB software, the surface layer porosity is defined as 0, fine aggregate as 1, asphalt phase as 2, base layer fine aggregate as 3, and cement phase as 4. The volume fraction of each phase material in each layer can be defined by equation (1-1): (1-1) In the formula, It represents the pore volume fraction. This represents the volume fraction of the asphalt phase. This represents the volume fraction of the cement phase.

[0007] Run MATLAB software to create hdf5 files containing arrays 0, 1, 2, 3, and 4, randomly distributed by volume fraction, and create txt files containing the dielectric properties of each phase material. The exported sphere coordinates and radius are used in gprmax software to reconstruct the coarse aggregate using the object construction command #sphere, and the dielectric properties of the coarse aggregate are defined using the material command #materials. #sphere: f1 f2 f3 f4 str1 f1, f2, and f3 represent the x, y, and z coordinates of each small ball, f4 represents the radius of each small ball, and str1 represents the name of the coarse aggregate.

[0008] #material: f1 f2 f3 f4 str1 f1, f2, f3, and f4 represent the dielectric constant, electrical conductivity, magnetic permeability, and magnetic loss of the coarse aggregate, respectively, and str1 represents the name of the coarse aggregate.

[0009] The code #geometry_objects_write is used to convert the coarse aggregate information from the secondary reconstruction into an hdf5 file, thereby improving the efficiency of the electromagnetic model in reading a large number of small balls; The HDF5 files of coarse aggregate phase and random medium phase are written into the gprmax software using the reading code #geometry_objects_read; By setting parameters such as time window, antenna frequency, and moving step size, a three-dimensional multiphase heterogeneous electromagnetic differential model is established. The model is in .in file format. Running the file in gprmax software can obtain the echo response under the three-dimensional multiphase heterogeneous model.

[0010] According to another aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the method for constructing a multiphase heterogeneous model for three-dimensional ground-penetrating radar forward modeling of the present invention.

[0011] According to another aspect of the present invention, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method for constructing a multiphase heterogeneous model for three-dimensional ground-penetrating radar forward modeling according to the present invention.

[0012] Compared with existing technologies, the beneficial effects of the above-described method of the present invention are as follows: 1. This invention reconstructs the true shape of coarse aggregates in road materials based on discrete element theory, and distributes the dielectric properties of irregular shapes such as pores and mortar based on random medium theory, making the model calculation results more realistic; 2. This invention is a three-dimensional, multiphase, heterogeneous electromagnetic differential model that can obtain the full-profile spectrum characteristics of highway structural defects echoes under heterogeneous interference. Attached Figure Description

[0013] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a pavement coarse aggregate reconstruction model obtained in a preferred embodiment of the present invention; Figure 2 This is a random medium model of each layer of the road surface obtained in a preferred embodiment of the present invention; Figure 3 This is a three-dimensional multiphase heterogeneous model of highway asphalt pavement obtained in a preferred embodiment of the present invention; Figure 4 These are the radar spectrum features under heterogeneous interference output in the preferred embodiment of the present invention; Figure 5 This is a flowchart of a preferred embodiment of the method provided by the present invention. Detailed Implementation

[0014] 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.

[0015] 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.

[0016] like Figure 1-5 As shown, a method for constructing a multiphase heterogeneous model for three-dimensional ground-penetrating radar forward modeling includes the following steps: investigating typical structures of highway asphalt pavement; setting the size and network of the multiphase heterogeneous model; calculating the coarse aggregate gradation curves of the surface layer and base layer; reconstructing the morphology of coarse aggregate and deriving the coordinates and radius of the small spheres; calculating the volume fraction of fine aggregate, pores, and asphalt; creating an HDF5 array and dielectric property file for the random phase; reconstructing the coarse aggregate in gprmax; creating an HDF5 array and dielectric property file for the coarse aggregate; and establishing a three-dimensional multiphase heterogeneous electromagnetic model.

[0017] Given that ground-penetrating radar prior maps are obtained based on forward modeling technology, the electromagnetic wave response calculated in the simulation differs significantly from the measured data because the forward modeling model fails to fully measure the heterogeneous characteristics of the pavement material. Therefore, this application provides a method for constructing a three-dimensional multiphase heterogeneous model of highway asphalt pavement based on the coupling of discrete element reconstruction and random medium theory. This method realizes the random distribution of coarse aggregate morphology and composite phase dielectric properties, thereby improving the authenticity and accuracy of the forward modeling.

[0018] The following details the specific steps of the method for constructing a three-dimensional multiphase heterogeneous model of highway asphalt pavement based on the coupling of discrete element reconstruction and stochastic medium theory provided in this application: Step S1: A typical structure of highway asphalt pavement consists of three layers: SMA-13 ​​upper layer, AC-20 middle layer, and AC-25 lower layer, with thicknesses of 4cm, 6cm, and 8cm respectively, for a total thickness of 18cm. There are two base layers: cement-stabilized crushed stone upper base and cement-stabilized crushed stone lower base, both with a thickness of 18cm, for a total thickness of 36cm. Step S2: Based on the above highway structure composition, determine the model size to be 0.5m. 0.6m (3cm subgrade + 36cm base course + 18cm surface course + 3cm air layer) The current 3D ground-penetrating radar detection accuracy is 5mm, with a model mesh size of 0.002. (0.2m) 0.002 0.002; Step S3: According to the Technical Specification for Construction of Highway Asphalt Pavement (JTG F40-2004), calculate the gradation curves of SMA-13 ​​surface layer, AC-20 intermediate layer, and AC-25 bottom layer based on the median gradation value. The asphalt mixture is divided into coarse and fine aggregates with a particle size of 2.36mm. Based on the gradation ratio of coarse and fine aggregates, determine the gradation curve of coarse aggregate in each layer of the surface layer. The calculation results are shown in Table 4.

[0019] Table 1: Aggregate Gradation Range of Mixtures

[0020] Table 2: Median Aggregate Gradation of Mixed Materials

[0021] Table 3: Aggregate Gradation Curve of Mixture

[0022] Table 4: Coarse Aggregate Gradation Curve

[0023] Step S4: Calculate the gradation curve of cement-stabilized crushed stone base course according to the recommended gradation in the Technical Specification for Construction of Highway Pavement Base Course (JTGT F20-2015). The cement mixture is divided into coarse and fine aggregates with a particle size of 4.75mm. Based on the gradation ratio of coarse and fine aggregates, determine the gradation curve of coarse aggregate in cement-stabilized crushed stone base course. The calculation results are shown in Table 7.

[0024] Table 5: Median Aggregate Gradation of Mixed Materials

[0025] Table 6: Aggregate Gradation Curve of Mixture

[0026] Table 7: Coarse Aggregate Gradation Curve

[0027] Step S5: Based on the gradation curves of coarse aggregates in each layer, generate a coarse aggregate reconstruction model in PFC 3D software, as shown in the attached figure. Figure 1 As shown, the coordinates and radii of the spheres that make up the coarse aggregate are derived; Step S6: Set the porosity of the SMA-13 ​​top layer, AC-20 middle layer, and AC-25 bottom layer to 4%, 4.5%, and 4.5%, respectively; the asphalt-aggregate ratio to 5.7%, 4.5%, and 4.5%, respectively; set the aggregate mass to 100g and the aggregate density to 2.7g / cm³. 3 The density of asphalt is 1.02 g / cm³. 3 The volume fractions of fine aggregate phase, porous phase and asphalt phase were calculated, as shown in Table 8. Table 8: Volume percentage of each phase in random medium

[0028] Step S7: Set the porosity and water-cement ratio of the upper and lower base layers to 2% and 6% respectively, set the aggregate mass to 100g, and the aggregate density to 2.7g / cm³. 3 The density of the cement is 1.02 g / cm³. 3 The volume fractions of fine aggregate phase, porous phase and cement phase were calculated, as shown in Table 9. Table 9: Volume percentage of each phase in random medium

[0029] Step S8: In MATLAB software, define the surface layer porosity as 0, fine aggregate as 1, asphalt phase as 2, base layer fine aggregate as 3, and cement phase as 4. The volume fraction of each phase material in each layer can be defined by equation (1-1): (1-1) In the formula, It represents the pore volume fraction. 1 represents the volume fraction of the asphalt phase, and 2 represents the volume fraction of the cement phase. For the SMA-13 ​​top layer, it is 0.119 in Table 8 and 0.373 in Table 8. For the cement-stabilized crushed stone base layer, it is 0.052 in Table 9 and 0.351 in Table 8.

[0030] Step S9: Run MATLAB software to randomly distribute and store arrays 0, 1, 2, 3, and 4 according to volume fraction in an hdf5 file. Create a txt file containing the dielectric properties of each phase material. The information in the txt file is as follows: #material: 1 0 1 0 0 #material: 8 1e-05 1 0 1 #material: 2.6 0.001 1 0 2 #material: 10 0.01 1 0 3 #material: 4 0.001 1 0 4 The generated stochastic medium model is attached. Figure 2 As shown; Step 10: In step S4, the exported sphere coordinates and radius are used in gprmax software to reconstruct the coarse aggregate using the object construction command #sphere. The surface coarse aggregate is named pebble, and the base coarse aggregate is named pebble1. Some definitions are as follows: The dielectric properties of the coarse aggregate are defined using the material command #materials. Since the base layer is more moist, the dielectric constant and conductivity of the base coarse aggregate are greater than those of the surface coarse aggregate, as defined below: #sphere: 0.998949 0.568467 0.0169173 0.00114208 pebble #sphere: 1.00049 0.38332 0.0156268 0.00175283 pebble1 #material: 8 1e-05 1 0 pebble #material: 10 0.01 1 0 pebble1; Step 11: Use the overwrite code #geometry_objects_write to convert the coarse aggregate information from the secondary reconstruction into an hdf5 file to improve the efficiency of the electromagnetic model in reading a large number of small balls; #geometry_objects_write: 0 0.03 0 0.5 0.57 0.2 pavement Step 12: Use the reading code #geometry_objects_read to write the HDF5 files of the coarse aggregate phase and the random medium phase into the gprmax software; #geometry_objects_read: 0 0.03 0 XX\pavement.h5 XX\pavement_materials.txt Step 13: Set the time window to 20ns, antenna frequency to 800MHz, movement step size to 0.01m, antenna height to 0.57m, and antenna spacing to 5cm. Use the #box command to generate a 3cm soil base and a 3cm air layer at the top and bottom of the model to establish a three-dimensional multiphase heterogeneous electromagnetic differential model, as shown in the attached figure. Figure 3 As shown; the model format is an .in file. The echo response of the three-dimensional multiphase heterogeneous model is calculated using the gprmax software. (See attached image.) Figure 4 As shown.

[0031] The examples described herein are merely preferred embodiments of the invention and are not intended to limit the concept and scope of the invention. Any modifications and improvements made by those skilled in the art to the technical solutions of the invention without departing from the design concept of the invention should fall within the protection scope of the invention.

Claims

1. A method for constructing a multiphase heterogeneous model for three-dimensional ground penetrating radar forward modeling, characterized in that: The steps include: Investigate the typical structure of asphalt pavement on expressways; Setting the size and network of multiphase heterogeneous models; Calculate the gradation curves of coarse and urgent materials of the surface layer and base layer; Reconstruct the coarse aggregate morphology and derive the sphere coordinates and radius; Calculate the volume fractions of fine aggregate, voids, and asphalt; Create hdf5 arrays and dielectric properties files of random phases; Secondary reconstruction of rough material in gprmax; Create hdf5 array and dielectric property file of coarse aggregate; A three-dimensional multiphase inhomogeneous electromagnetic analysis model is established.

2. The method according to claim 1, characterized in that The typical structure of asphalt pavement of highways investigated is as follows: according to the typical structure of asphalt pavement of highways, the surface layer is composed of S layer, each paving material is MS1, MS2, MS3, and the thickness of each layer of the surface layer is HS1, HS2, HS3; the base layer is composed of B layer, each paving material is BS1, BS2, and the thickness of each layer of the surface layer is HS1, HS2.

3. The method according to claim 2, characterized in that The size and network of the multiphase heterogeneous model are set as follows: the model size is determined according to the typical structural composition of the asphalt pavement of the highway, and the model grid size is set according to the current three-dimensional ground penetrating radar detection accuracy.

4. The method according to claim 3, characterized in that The specific steps for calculating the grading curves of coarse aggregates for the surface layer and the base layer are as follows: Calculate the grading curves of the paving materials MS1, MS2, and MS3 for each layer of the asphalt surface layer; divide the asphalt mixture into coarse and fine aggregates based on a particle size of 2.36 mm; determine the grading curves of the coarse aggregates for each layer of the surface layer according to the gradation ratio of the coarse and fine aggregates; Calculate the grading curves of the paving materials BS1 and BS2 for each layer of the base layer; divide the cement mixture into coarse and fine aggregates based on a particle size of 4.75 mm; determine the grading curves of the coarse aggregates for each layer of the base layer according to the gradation ratio of the coarse and fine aggregates.

5. The method according to claim 4, characterized in that Reconstruct the coarse aggregate morphology and derive the coordinates and radius of the small balls as follows: According to the grading curve of the coarse aggregate in each layer, generate the coarse aggregate reconstruction model in the PFC 3D software, and derive the coordinates and radius of the small balls that make up the coarse aggregate.

6. The method according to claim 1, characterized in that The three-dimensional multiphase inhomogeneous electromagnetic differential model is established as follows: parameters are set to establish a three-dimensional multiphase inhomogeneous electromagnetic differential model; the model format is an in file, and the file is run in the gprmax software to obtain the echo response under the three-dimensional multiphase inhomogeneous model.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps in the method for constructing a multiphase inhomogeneous model for three-dimensional ground penetrating radar forward modeling as described in any one of claims 1 to 6 are implemented.

8. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps in the method for constructing a multiphase inhomogeneous model for three-dimensional ground penetrating radar forward modeling as described in any one of claims 1 to 6 are implemented.

Images