A method for judging the degree of sinking damage in heavy-haul railway ballast based on ground penetrating radar

Abstract

The present invention discloses a method for judging the degree of damage of heavy-duty railway ballast groove based on ground penetrating radar, which belongs to the field of ground penetrating radar data processing technology, and includes the following steps: obtaining ground penetrating radar data of heavy-duty railway and preprocessing; layering the preprocessed ground penetrating radar data; based on the wave field reverse continuation principle, offsetting and merging the layered ground penetrating radar data by introducing data matrices of different layers to obtain complete damage offset data; based on the complete damage offset data, obtaining damage feature data by calculation; constructing a functional relationship for judging the degree of damage of heavy-duty railway ballast groove, and obtaining the degree of damage of heavy-duty railway ballast groove based on the functional relationship and the damage feature data. The method proposed by the present invention can more accurately judge the severity of the damage of heavy-duty railway ballast groove.

Description

Technical Field

[0001] The invention belongs to the technical field of ground penetrating radar data processing, and in particular relates to a method for judging the degree of heavy-load railway ballast groove damage based on ground penetrating radar. Background Art

[0002] The roadbed is crucial to heavy-haul railways and train safety. Due to the large load and long body length of heavy-haul trains, the ballast acts on the same position of the line for a long time. As the train density continues to increase, the ballast will gradually sink into the roadbed, forming the most common disease in heavy-haul railways - ballast grooving. For the track structure of the entire heavy-haul railway, after the ballast grooving disease occurs, the track unevenness and track sinking will be further aggravated until the stability and safety of train operation are severely challenged.

[0003] At present, most of the judgments on the severity of heavy-duty railway ballast groove defects are made by manual experience. This method has a low accuracy rate and is prone to secondary damage to the roadbed. Ground penetrating radar nondestructive testing technology is widely used in heavy-duty railway roadbed defect detection projects, but it is basically carried out by offset processing. Commonly used offset technologies such as FK offset have certain shortcomings. The commonly used FK offset is more suitable for offset processing of defects in single-layer uniform media, while heavy-duty railway ballast groove defects are usually multi-layered, resulting in the commonly used FK offset having to calculate the wave velocity based on the equivalent dielectric constant of the multi-media layer where the defect is located, and thus it is impossible to perform good offset processing on defects that appear on multiple layers, and it is also impossible to detect the defects of heavy-duty railway ballast grooves.

[0004] Therefore, it is urgent to propose a method to determine the degree of heavy-duty railway ballast groove damage based on ground penetrating radar. Summary of the invention

[0005] In order to solve the above technical problems, the present invention proposes a method for determining the degree of damage of heavy-duty railway ballast grooves based on ground penetrating radar to solve the problems existing in the above prior art.

[0006] To achieve the above object, the present invention provides a method for determining the degree of damage of heavy-duty railway ballast grooves based on ground penetrating radar, comprising the following steps:

[0007] Acquire and pre-process ground penetrating radar data of heavy-haul railways;

[0008] Performing layered processing on the preprocessed ground penetrating radar data;

[0009] Based on the principle of wave field reverse continuation, the layered GPR data are migrated and merged by introducing different layered data matrices to obtain complete disease migration data.

[0010] Based on the complete defect offset data, defect characteristic data is obtained by calculation, wherein the defect characteristic data includes the thickness, upper top surface width and lower bottom surface width of the ballast groove defect;

[0011] A functional relationship for determining the degree of damage of heavy-duty railway ballast grooves is constructed, and the degree of damage of heavy-duty railway ballast grooves is obtained based on the functional relationship and the damage characteristic data.

[0012] Optionally, the process of preprocessing the ground penetrating radar data of the heavy-haul railway includes:

[0013] The direct wave removal and exponential gain processing are performed on the ground penetrating radar data of heavy-haul railway to obtain the preprocessed ground penetrating radar data.

[0014] Optionally, the process of layering the preprocessed ground penetrating radar data includes:

[0015] Each column of the preprocessed ground penetrating radar data is cumulatively summed, and stratified based on the fluctuation range of the cumulatively summed data to obtain stratified ground penetrating radar data;

[0016] The layered ground penetrating radar data includes an air layer, a ballast layer and a road base layer in sequence.

[0017] Optionally, the process of performing migration processing on the layered ground penetrating radar data includes:

[0018] Perform FK migration on the medium in the air layer to obtain the air layer migration result;

[0019] Acquire the first layer of the air layer and the ballast layer, and the second layer of the ballast layer and the roadbed layer based on the accumulated amplitude of the ground penetrating radar;

[0020] For the medium in the ballast layer, the wave field data received by the ground penetrating radar is firstly subjected to Fourier transformation, thereby obtaining wave field data in the frequency domain and wave number domain; then, based on the wave field reverse continuation principle, the first data matrix is ​​introduced to perform electromagnetic wave reverse continuation offset phase compensation, thereby obtaining wave field data at the first layer; the wave field data at the first layer is subjected to two-dimensional inverse Fourier transformation, thereby obtaining wave field data at the first layer in the time domain; the detection relationship of the ground penetrating radar is obtained, and based on the detection relationship of the ground penetrating radar and the wave field data at the first layer in the time domain, the ballast layer offset result is obtained;

[0021] For the medium in the roadbed layer, the wave field data received by the ground penetrating radar is first Fourier transformed to obtain the wave field data in the frequency domain and wave number domain; then, based on the wave field reverse continuation principle, the second data matrix is ​​introduced to perform electromagnetic wave reverse continuation offset phase compensation to obtain the wave field data at the second layer; the wave field data at the second layer is subjected to a two-dimensional inverse Fourier transform to obtain the wave field data at the second layer in the time domain; the detection relationship of the ground penetrating radar is obtained, and based on the detection relationship of the ground penetrating radar and the wave field data at the second layer in the time domain, the roadbed layer offset result is obtained.

[0022] Optionally, the first data matrix is ​​as follows:

[0023] ,

[0024] in, is the wave number component in the moving direction of the ground penetrating radar antenna, is the angular frequency, h is the thickness of the first layer, v1 is the equivalent wave velocity of the air layer medium, k z is the wave number component in the detection direction of the ground penetrating radar, and j is an imaginary unit.

[0025] Optionally, the detection relationship of the ground penetrating radar is as follows:

[0026] ,

[0027] in, is the wave number component in the moving direction of the ground penetrating radar antenna, is the wave number component in the detection direction of the ground penetrating radar, is the angular frequency.

[0028] Optionally, based on the complete disease deviation data, the process of obtaining disease characteristic data by calculation includes:

[0029] The complete disease offset data is converted into a ground penetrating radar two-dimensional image and preprocessed. The preprocessed ground penetrating radar two-dimensional image is divided based on the maximum inter-class variance method to obtain a grayscale image divided into a background area and a foreground area; the grayscale image is binarized, and then the Canny edge detection algorithm is used to extract the edge of the binarized image to obtain an edge image of the foreground area; the upper top surface width and the lower bottom surface width are obtained based on the edge image of the foreground image.

[0030] Optionally, the process of obtaining the disease characteristic data by calculation based on the complete disease deviation data further includes:

[0031] Based on the quantitative relationship between the time difference of the reflected waves on the upper surface and the lower surface, the thickness of the ballast groove defect is obtained;

[0032] The calculation formula for the thickness of the ballast groove disease is as follows:

[0033] ,

[0034] Among them, T1 is the time corresponding to the reflection wave of the ballast layer and the roadbed layer, T2 is the time corresponding to the reflection wave of the bottom surface under the ballast groove, z is the thickness of the ballast groove, b is the distance between the transmitting and receiving antennas of the ground penetrating radar, and c is the speed of electromagnetic wave propagation in vacuum. is the dielectric constant of the substrate.

[0035] Optionally, the functional relationship for judging the degree of heavy-duty railway ballast groove damage is as follows:

[0036] ,

[0037] Among them, S is the severity score of heavy-duty railway ballast groove disease, W1 is the width of the upper surface of the groove, W2 is the width of the lower surface of the groove, and H is the thickness of the ballast groove disease.

[0038] Optionally, the process of obtaining the damage degree of the heavy-load railway ballast groove further includes:

[0039] A damage degree classification standard is preset, and the severity of heavy-duty railway ballast grooves is obtained based on the judgment formula for the damage degree of heavy-duty railway ballast grooves and the damage degree classification standard.

[0040] Compared with the prior art, the present invention has the following advantages and technical effects:

[0041] The present invention provides a method for determining the degree of damage of heavy-duty railway ballast grooves based on ground penetrating radar.

[0042] The FK migration is improved, and based on the principle of reverse continuation of the wave field, a new data matrix is ​​introduced when processing each layer, the position of the ground penetrating radar receiving antenna is changed, and the migration processing is performed in sequence from top to bottom according to the layered data, so that the multi-layer disease migration processing results can be obtained. Compared with the prior art that can only realize the disease migration processing of a single-layer medium, the present invention proposes a better migration processing data method; according to the obtained migration processing data, the size of the disease feature of the heavy-duty railway ballast groove can be more accurately calculated, especially for the calculation of the width of the top surface depression of the more complex ballast groove; according to the calculated size of the disease feature of the heavy-duty railway ballast groove, combined with the given judgment formula for the severity of the heavy-duty railway ballast groove, the severity of the disease can be more accurately judged; at the same time, the method proposed by the present invention has low complexity, fast calculation processing speed, and requires very little computing power, is more suitable for outdoor simple environment detection, and has a wide range of application scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0044] Figure 1 is a flow chart of a method according to an embodiment of the present invention;

[0045] Figure 2 A schematic diagram of a heavy-duty railway ballast groove disease model according to an embodiment of the present invention;

[0046] Figure 3 This is a schematic diagram of the hierarchical principle of preprocessing data according to an embodiment of the present invention;

[0047] Figure 4 A comparison diagram of complete disease migration image results of an embodiment of the present invention;

[0048] Figure 5 A schematic diagram of the edge of a ballast groove after using the maximum inter-class variance method and the Canny algorithm according to an embodiment of the present invention;

[0049] Figure 6 A comparison diagram of the calculation of the width of the depression on the top surface of the heavy-duty railway ballast groove according to an embodiment of the present invention;

[0050] Figure 7 A comparison diagram of the error rate calculation of the width of the depression on the top surface of the heavy-duty railway ballast groove according to an embodiment of the present invention;

[0051] Figure 8 This is a characteristic diagram of ballast groove damage in an embodiment of the present invention. DETAILED DESCRIPTION

[0052] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0053] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0054] Embodiment 1

[0055] like Figure 1As shown, in this embodiment, a method for judging the degree of damage of heavy-duty railway ballast groove based on ground penetrating radar is provided, comprising the following steps: obtaining ground penetrating radar data of heavy-duty railway and performing preprocessing, including removing interference of direct waves, gain and data stratification; then performing two-dimensional Fourier transform on the stratified data from top to bottom, converting it into FK domain for analysis; converting the analyzed data into time domain and then merging it to obtain complete offset processed data; using the maximum inter-class variance method to divide the complete offset data image after preprocessing by gray value, and then extracting the damage characteristics by calculating the gradient component along the gradient direction based on the Canny edge detection algorithm, and extracting the track number where the top surface depression and the bottom surface edge of the ballast groove are located by setting a threshold, so as to improve the extraction accuracy; then calculating its thickness characteristics by the time difference of the reflected wave of the top surface depression and the bottom surface of the ballast groove, and then making a more accurate judgment on the severity of the damage of the heavy-duty railway ballast groove according to the given formula for judging the severity of the damage.

[0056] As a specific embodiment, the method comprises the following steps:

[0057] Step 1: Get the raw ground penetrating radar image data:

[0058] Based on the actual situation of heavy-haul railway trackbed, the FDTD-based gprMax software is used to perform forward simulation of the ballast groove disease in the trackbed to generate ground penetrating radar image data.

[0059] It is feasible to use the ballast commonly used in heavy-duty railways as the main component of the track bed model. The model is composed of air layer, ballast, and base layer from top to bottom. The thickness of the air layer is set to 40cm, and the relative dielectric constant is 1; the thickness of the ballast layer is set to 40cm, and the relative dielectric constant is 5; the thickness of the base layer is set to 80cm, and the relative dielectric constant is 20. Ballast grooves mainly occur at the interface between the ballast layer and the base layer, caused by the ballast layer sinking into the base layer, so the top surface width of the ballast groove disease is set to 60cm, the bottom surface width is 30cm, and the thickness is 40cm. In general, the conductivity of the heavy-duty railway track bed is extremely low, non-ferromagnetic and relatively dry, so the conductivity is 0 and the relative magnetic permeability is 1. The heavy-duty railway ballast groove disease model is as follows: Figure 2 The center frequency of the radar antenna in the forward simulation was set to 1500MHz, the number of channels was 61, the interval was 0.025m, and the number of sampling points was 6361. The simulation obtained the ground penetrating radar ballast groove disease data image, which was displayed in the form of B-scan.

[0060] Step 2: Ground Penetrating Radar Data Preprocessing:

[0061] The ground penetrating radar image data obtained based on simulation is processed by removing direct waves and exponential gain to obtain corresponding pre-processed two-dimensional image data.

[0062] (1) Ground penetrating radar direct wave removal processing:

[0063] Since the direct wave signal has strong energy and tends to exist stably, it will hide the real ballast groove defect signal. Since the direct wave and the effective signal are mixed together and difficult to separate in the time domain, the signal is converted to the frequency-wave number domain through Fourier transform. In this domain, the direct wave can be considered to be horizontally coherent, and its energy is mainly concentrated at the position where the wave number is zero. Based on this feature, a fan-shaped filter can be used to effectively filter out the direct wave. The specific process is as follows:

[0064] (1)

[0065] The data collected by the ground penetrating radar is w(x, t), where x is the moving direction of the ground penetrating radar survey line, t is the time, u(x, t) is the direct wave, and z(x, t) is the effective ground penetrating radar signal.

[0066] After Fourier transform, the frequency domain can be expressed as:

[0067] (2)

[0068] In the time domain, the direct wave and the effective GPR signal are superimposed and cannot be effectively separated. In the case of a flat detection, the direct wave is mainly concentrated in k x = 0, perform two-dimensional directed wavelet transform on the effective ground penetrating radar signal, select appropriate anisotropic parameters, and then get the approximate value of the direct wave. Then, through formula (1), get the effective ground penetrating radar signal.

[0069] (2) GPR data exponential gain processing:

[0070] The high-frequency electromagnetic signal emitted by the ground-penetrating radar will be refracted or reflected when propagating in different media of the heavy-duty railway track bed, resulting in the loss of electromagnetic wave energy. The ground-penetrating radar data is amplified by exponential gain.

[0071] The gain processing of ground penetrating radar is generally defined as:

[0072] (3)

[0073] Among them, x(n) is the original data of the ground penetrating radar, y(n) is the data after the ground penetrating radar gain, and f(n) is the gain function.

[0074] (4)

[0075] Where a is the linear gain coefficient and b is the exponential gain coefficient.

[0076] Step 3: Layering of preprocessed data:

[0077] Based on the preprocessed ground penetrating radar two-dimensional image data, each channel of ground penetrating radar data is accumulated and summed, and then layered according to the fluctuation range of the accumulated and summed data to obtain layered image data.

[0078] It is feasible to accumulate and sum each column of the preprocessed ground penetrating radar two-dimensional image data, calibrate the sampling point range when the amplitude is greater than 100 according to the data fluctuation range after the accumulation and summation, and calibrate the maximum boundary value of the sampling point when the amplitude is greater than 100 as the critical point for the preprocessed data stratification. The preprocessed data is divided into three layers, where the first layer is mainly the air layer, the second layer is mainly the ballast layer, and the third layer is mainly the roadbed layer. For details, see Figure 3 .

[0079] Step 4: Layered offset processing:

[0080] Based on the layered image data, a two-dimensional Fourier transform is performed to convert it to the FK domain. Then, by importing a new data matrix, the position of the ground penetrating radar receiving antenna is changed, and the offset processing is performed from top to bottom according to the layered data to obtain the offset processed data.

[0081] The feasible calculation of layered migration is mainly to study the migration calculation of the second and third layers of media except the first layer. The first layer of media from top to bottom is still suitable for normal FK migration for imaging. The h and h1 at the media stratification are found by accumulating the amplitude of the ground penetrating radar. The thickness h of the first layer of media is used as a known parameter for the study of the second layer migration calculation.

[0082] According to Snell's law, when there is a discontinuity in the electromagnetic properties of a substance, the electromagnetic wave will be refracted. Therefore, in this case, the electromagnetic wave no longer propagates in a straight line and at a constant speed to reach the receiving antenna, but is divided into two segments. Therefore, in a uniform and isotropic medium, the wave equation is no longer applicable. In this case, it can be considered that the electromagnetic wave propagates in the first and second layers of the medium, and the wave equation applies to these two media respectively.

[0083] Therefore, two wave equations can be used to represent these two homogeneous media:

[0084] (5)

[0085] Among them, x is the horizontal distance of the ground penetrating radar, z is the measurement depth direction of the ground penetrating radar, v1 is the equivalent wave velocity of the first layer of medium, which is half of the true wave velocity, and v2 is the equivalent wave velocity of the second layer of medium, which is half of the true wave velocity.

[0086] In the first layer of medium, , and The relationship is:

[0087] (6)

[0088] in, is the wave number component in the moving direction of the ground penetrating radar antenna, is the wave number component in the detection direction of the ground penetrating radar, is the angular frequency of t.

[0089] The main research content of layered migration calculation is to use the wave field received when z=0 According to the principle of reverse extension of wave field, the new data matrix A is imported through the data calculation after model amplitude stratification, and the phase compensation in the frequency wavenumber domain is carried out to deduce the assumed GPR receiving antenna in GPR wave field Then, the normal FK offset of the single-layer medium is used for imaging again. The specific principle is as follows:

[0090] First, the wave field received by the ground penetrating radar Perform Fourier transform, and then we can get the frequency domain and wave number domain , and then according to the principle of reverse extension of the wave field, the wave field value extended to the layering h is:

[0091] (7)

[0092] in,

[0093] (8)

[0094] The data matrix A is used to calculate the electromagnetic wave reverse extension offset phase compensation in the layered offset calculation and can be calculated by formula (5).

[0095] k on both sides of the above equation x and Perform a two-dimensional inverse Fourier transform, and then you can get the wave field at time t and depth z in the time domain, that is:

[0096] (9)

[0097] From the explosion reflection interface model, we know that the desired wave field is at the moment t=0, that is, when t=0 in equation (9), we can get:

[0098] (10)

[0099] By transforming formula (6), we can get:

[0100] (11)

[0101] Substitute equation (11) into equation (10), and then use the interpolation method to transform it into the time domain, and finally obtain the imaging result:

[0102] (12)

[0103] The imaging principle of the third layer of medium is the same as the offset principle of the second layer of medium. It is only necessary to replace h in the above formula with the thickness of the second layer of medium, and then the data after the first, second and third layers are offset can be obtained. Here, the data matrix in the offset principle of the second layer of medium is called the first data matrix, and the data matrix in the offset principle of the third layer of medium is called the second data matrix.

[0104] Step 5: Merging of disease offset data:

[0105] Based on the data processed by the offset, they are merged in sequence according to the order of offset to obtain complete disease offset data.

[0106] Available, see Figure 4 ,Based on the data processed by the offset, they are merged in sequence according to the order of offset to obtain the complete disease offset data.

[0107] Step 6: Calculate disease characteristic data:

[0108] Based on the complete defect offset data, the thickness of the ballast groove defect, the width of the depression on the upper top surface and the lower bottom surface are calculated to obtain the defect characteristic data. At the same time, in order to better verify that the improved technology can well calculate the width of the depression on the upper top surface of the ballast groove defect, a large number of target simulation verifications were carried out.

[0109] It is feasible to convert the complete disease migration data into a ground penetrating radar two-dimensional Bscan image, and use the maximum inter-class variance method to divide the preprocessed ground penetrating radar two-dimensional Bscan image by gray value. The histogram of the grayscale image is obtained by the imhist function, that is, the number and distribution of pixels at each gray level. Assuming that the grayscale range of the grayscale image is [0,255], the number of pixels at each gray level is counted. By defining the initial values ​​of the current maximum inter-class variance currentMax and the optimal threshold optimalThreshold. sumTotal represents the total grayscale value of the entire image, which is used to calculate the overall mean. The maximum inter-class variance method is used to traverse each possible grayscale level t as the threshold. For each threshold t, the image is divided into two parts: background (grayscale value in [0,t−1]) and foreground (grayscale value in [t,255]), and the following parameters are calculated:

[0110]

[0111] in is the ratio of background pixels, is the proportion of foreground pixels, is the grayscale mean in the background area, is the grayscale mean in the foreground area, is the inter-class variance under the current threshold.

[0112] According to the optimal threshold optimalThreshold, the grayscale image is binarized, that is, the pixel values ​​above the threshold are set to foreground white, and the pixel values ​​below the threshold are set to background black.

[0113] The Canny edge detection algorithm is applied to the binary image. This method can further extract the detailed boundary information of the ballast groove disease characteristics in the foreground area and obtain an accurate edge image, such as Figure 5 As shown. Then the number of the track where the boundary is located is accurately read, and the interval of the track number occupied by the disease characteristics is extracted.

[0114] Calculation shows that the total number of abnormal intervals on the top surface of the ballast groove defect is 25, and the total number of abnormal intervals on the bottom surface is 12. According to the interval of 2.50 cm set for the ground penetrating radar during simulation, it is calculated that the width of the depression on the top surface of the ballast groove defect is 62.50 cm, and the width on the bottom surface is 30.00 cm. Compared with the width of the top surface and the width of the bottom surface of the defect set in the simulation, the calculation error of the width of the top surface depression is 4.17%, and the calculation error of the width of the bottom surface is 0%.

[0115] In order to better verify that the improved offset technology can be better applied to the calculation of the width of the depression on the top surface of heavy-duty railway ballast groove disease, six groups of verification models were established again using gprMax. The overall size of the model is 9.9×1.6m, in which ballast groove diseases with a depression width of 35cm, 40cm, 45cm, 50cm, 55cm, and 65cm on the top surface were set respectively. From left to right, they are target one, target two, target three, target four, target five, and target six. The remaining parameters of the model and radar forward simulation are the same as those of the heavy-duty railway ballast groove disease model, and the width of the depression on the top surface based on the improved offset technology is calculated. At the same time, the same method is used to extract the abnormal track number interval and calculate the results of the data without any processing and normal FK offset, and the width and error size are compared. As shown in Table 1, Figure 6 and Figure 7 shown.

[0116] Table 1

[0117] Based on the complete offset data of the ballast groove model and the quantitative relationship between the thickness of the ballast groove and the time difference of the reflected waves above and below the ballast groove, the calculation formula for the thickness of the ballast groove defect can be given, namely:

[0118] (14)

[0119] Among them, T1 is the time corresponding to the reflection wave of the ballast layer and the roadbed layer, T2 is the time corresponding to the reflection wave of the bottom surface under the ballast groove, z is the thickness of the ballast groove, b is the distance between the transmitting and receiving antennas of the ground penetrating radar, and c is the speed of electromagnetic wave propagation in vacuum. is the dielectric constant of the substrate.

[0120] According to the complete offset data and the calculation formula of the ballast groove thickness, the thickness of the ballast groove can be calculated to be 40.57 cm, and the calculation error with the thickness set by the simulation is 1.43%.

[0121] From the calculation results, it can be seen that the improved offset method is well suitable for the calculation of the characteristics of heavy-duty railway ballast groove disease, and the calculation error is generally small, and the characteristic data of ballast groove disease can be calculated more accurately.

[0122] Step 7: Determine the severity of the disease:

[0123] Based on the obtained ballast groove disease characteristic data and combined with the given heavy-duty railway ballast groove severity judgment formula, the severity of the ballast groove disease can be judged more accurately and quickly.

[0124] Available, see Figure 8 In the process of evaluating the severity of ballast groove disease, the three parameters of the upper top surface width, lower bottom surface width and thickness of heavy-duty railway ballast groove disease play a decisive role. In order to more intuitively and conveniently judge the severity of heavy-duty railway ballast groove disease, the following formula for judging the severity of heavy-duty railway ballast groove disease is given:

[0125] (16)

[0126] Among them, S is the severity score of heavy-duty railway ballast groove disease, W1 is the width of the upper surface of the groove / cm, W2 is the width of the lower surface of the groove / cm, and H is the thickness of the ballast groove disease / cm.

[0127] Since heavy-load railway trains have a large load, more attention is paid to the impact of defects on the bearing capacity of railway subgrade, especially the impact of the thickness of ballast groove defects on the safe operation of heavy-load railway trains. Therefore, the proportion of the thickness of ballast groove defects is increased in the judgment formula to more accurately judge the severity of heavy-load railway ballast groove defects. The classification standards for the severity of ballast groove defects and recommended treatment methods are given, as shown in Table 2.

[0128] Table 2

[0129] Substituting the previously calculated upper width of 62.5 cm, lower width of 30 cm, and thickness of 40.57 cm into formula (15), the severity score of heavy-duty railway ballast groove disease is calculated to be 42.84, and the grading standard of ballast groove disease severity is grade III. Therefore, according to the given calculation method of the upper top surface, lower bottom surface and thickness characteristics of heavy-duty railway ballast groove disease, as well as the disease severity judgment formula and grading standard, the severity of heavy-duty railway ballast groove disease can be judged more systematically and quickly.

[0130] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for determining the degree of damage to heavy-duty railway ballast grooves based on ground penetrating radar, characterized in that: The following steps are involved: Acquire and pre-process ground penetrating radar data of heavy-haul railways; Performing layered processing on the preprocessed ground penetrating radar data; Based on the principle of wave field reverse continuation, the layered GPR data are migrated and merged by introducing different layered data matrices to obtain complete disease migration data. Based on the complete defect offset data, defect characteristic data is obtained by calculation, wherein the defect characteristic data includes the thickness, upper top surface width and lower bottom surface width of the ballast groove defect; A functional relationship for determining the degree of damage of heavy-duty railway ballast grooves is constructed, and the degree of damage of heavy-duty railway ballast grooves is obtained based on the functional relationship and the damage characteristic data.

2. The method for determining the degree of heavy-duty railway ballast groove damage based on ground penetrating radar according to claim 1 is characterized in that: The process of preprocessing the GPR data of heavy-haul railway includes: The direct wave removal and exponential gain processing are performed on the ground penetrating radar data of heavy-haul railway to obtain the preprocessed ground penetrating radar data.

3. The method for determining the degree of heavy-duty railway ballast groove damage based on ground penetrating radar according to claim 1 is characterized in that: The process of layering the preprocessed GPR data includes: Each column of the preprocessed ground penetrating radar data is cumulatively summed, and stratified based on the fluctuation range of the cumulatively summed data to obtain stratified ground penetrating radar data; The layered ground penetrating radar data includes an air layer, a ballast layer and a road base layer in sequence.

4. The method for determining the degree of heavy-duty railway ballast groove damage based on ground penetrating radar according to claim 3 is characterized in that: Based on the principle of wave field reverse continuation, the process of migration processing of layered GPR data by introducing different layered data matrices includes: Perform FK migration on the medium in the air layer to obtain the air layer migration result; Acquire the first layer of the air layer and the ballast layer, and the second layer of the ballast layer and the roadbed layer based on the accumulated amplitude of the ground penetrating radar; For the medium in the ballast layer, the wave field data received by the ground penetrating radar is firstly subjected to Fourier transformation, thereby obtaining wave field data in the frequency domain and wave number domain; then, based on the wave field reverse continuation principle, the first data matrix is ​​introduced to perform electromagnetic wave reverse continuation offset phase compensation, thereby obtaining wave field data at the first layer; the wave field data at the first layer is subjected to two-dimensional inverse Fourier transformation, thereby obtaining wave field data at the first layer in the time domain; the detection relationship of the ground penetrating radar is obtained, and based on the detection relationship of the ground penetrating radar and the wave field data at the first layer in the time domain, the ballast layer offset result is obtained; For the medium in the roadbed layer, the wave field data received by the ground penetrating radar is first Fourier transformed to obtain the wave field data in the frequency domain and wave number domain; then, based on the wave field reverse continuation principle, the second data matrix is ​​introduced to perform electromagnetic wave reverse continuation offset phase compensation to obtain the wave field data at the second layer; the wave field data at the second layer is subjected to a two-dimensional inverse Fourier transform to obtain the wave field data at the second layer in the time domain; the detection relationship of the ground penetrating radar is obtained, and based on the detection relationship of the ground penetrating radar and the wave field data at the second layer in the time domain, the roadbed layer offset result is obtained.

5. The method for determining the degree of heavy-duty railway ballast groove damage based on ground penetrating radar according to claim 4 is characterized in that: The first data matrix is ​​as follows: ， in, is the wave number component in the moving direction of the ground penetrating radar antenna, is the angular frequency, h is the thickness of the first layer, v1 is the equivalent wave velocity of the air layer medium, k z is the wave number component in the detection direction of the ground penetrating radar, and j is an imaginary unit.

6. The method for determining the degree of heavy-duty railway ballast groove damage based on ground penetrating radar according to claim 4 is characterized in that: The detection relationship of the ground penetrating radar is as follows: ， in, is the wave number component in the moving direction of the ground penetrating radar antenna, is the wave number component in the detection direction of the ground penetrating radar, is the angular frequency.

7. The method for determining the degree of heavy-duty railway ballast groove damage based on ground penetrating radar according to claim 1, characterized in that: Based on the complete disease deviation data, the process of obtaining disease characteristic data by calculation includes: The complete disease offset data is converted into a ground penetrating radar two-dimensional image and preprocessed. The preprocessed ground penetrating radar two-dimensional image is divided based on the maximum inter-class variance method to obtain a grayscale image divided into a background area and a foreground area; the grayscale image is binarized, and then the Canny edge detection algorithm is used to extract the edge of the binarized image to obtain an edge image of the foreground area; the upper top surface width and the lower bottom surface width are obtained based on the edge image of the foreground area.

8. The method for determining the degree of heavy-duty railway ballast groove damage based on ground penetrating radar according to claim 7 is characterized in that: The process of obtaining the defect characteristic data by calculation based on the complete defect offset data further includes: obtaining the thickness of the ballast groove defect based on the quantitative relationship corresponding to the time difference of the reflected waves of the upper top surface and the lower bottom surface; The calculation formula for the thickness of the ballast groove disease is as follows: ， Among them, T1 is the time corresponding to the reflection wave of the ballast layer and the roadbed layer, T2 is the time corresponding to the reflection wave of the bottom surface under the ballast groove, z is the thickness of the ballast groove, b is the distance between the transmitting and receiving antennas of the ground penetrating radar, and c is the speed of electromagnetic wave propagation in vacuum. is the dielectric constant of the substrate.

9. The method for determining the degree of heavy-duty railway ballast groove damage based on ground penetrating radar according to claim 1, characterized in that: The functional relationship for judging the degree of heavy-duty railway ballast groove damage is as follows: ， Among them, S is the severity score of heavy-duty railway ballast groove disease, W1 is the width of the upper top surface of the groove, W2 is the width of the lower bottom surface of the groove, and H is the thickness of the ballast groove disease.

10. The method for determining the degree of heavy-duty railway ballast groove damage based on ground penetrating radar according to claim 9, characterized in that: The process of obtaining the damage degree of heavy-haul railway ballast sinking also includes: A damage degree classification standard is preset, and the severity of heavy-duty railway ballast grooves is obtained based on the judgment formula for the damage degree of heavy-duty railway ballast grooves and the damage degree classification standard.

Images