Track disengaging gap detection device and detection method

Abstract

The invention discloses a track disengaging gap detection device, the device comprises a data acquisition system, a data processing system and a transmission system, the data acquisition system is connected with the transmission system, the transmission system is connected with the data processing system, the data acquisition system is used for acquiring internal information of a ballastless track; the transmission system is used for information transmission between the data acquisition system and the data processing system; and the data processing system is used for processing the ballastless track internal information acquired by the data acquisition system to obtain processed information. In combination with a vector machine intelligent algorithm, accurate and efficient void disease detection is realized in a ballastless track disease detection process through ground penetrating radar initial detection, vector machine calculation of a suspected void gap position and shock wave method redetection. The method is excellent in performance in the field of roadbed engineering disease detection, high in detection efficiency, accurate in result and remarkable in application effect, and has wide application prospects and commercial value.

Description

Technical Field

[0001] The present invention relates to a ballastless railway disease detection device and a detection method, and in particular to a ballastless track gap detection device and a detection method. Background Art

[0002] Track debonding and cracking are common defects during subgrade track operation, directly reducing track dynamic smoothness and safety. Currently, crack detection relies primarily on manual inspection, which is costly, time-consuming, provides little information, and is prone to missing critical internal cracks. Effective rapid detection methods are also lacking. Summary of the Invention

[0003] The purpose of the present invention is to provide a track gap detection device and detection method, which combines the vector machine intelligent algorithm, uses ground penetrating radar for initial inspection, vector machine calculation of suspected gap locations, and shock wave method for re-inspection, to achieve accurate and efficient gap detection in the ballastless track defect detection process.

[0004] The purpose of the present invention is achieved through the following technical solutions:

[0005] A track gap detection device includes a data acquisition system, a data processing system and a transmission system, wherein the data acquisition system is connected to the transmission system, and the transmission system is connected to the data processing system, wherein:

[0006] The data acquisition system includes a positioning module, a radar detection module, and an impact echo supplementary detection module;

[0007] The positioning module is used to provide position information for the radar detection module;

[0008] The radar detection module is used to preliminarily collect internal information of the ballastless track;

[0009] The impact echo supplementary detection module is used to perform secondary detection on the suspected ballastless track points to determine the detailed information of the cracks;

[0010] The data processing system is used to process the internal information of the ballastless track collected by the data collection system to obtain processing information, including a first data plate, a gap preliminary analysis plate, a second data plate, and a gap confirmation analysis plate;

[0011] The first data block is used to collect and process ballastless track slab data obtained through preliminary detection by ground penetrating radar;

[0012] The crack preliminary analysis panel uses machine learning to determine the first data panel and is equipped with a positioning device to mark the location of suspected cracks.

[0013] The second data block is used to apply the impact echo detection method to the suspected crack position determined by the crack preliminary analysis block, re-detect the marked position, and transmit the detection data to the crack confirmation analysis block;

[0014] The crack confirmation and analysis section is used to process and confirm a series of indicators of crack location, width and shape.

[0015] A method for detecting track gaps and crevices comprises the following steps:

[0016] Step 1: Using a ground penetrating radar method, a gap detection is performed on the non-operated track to obtain first detection data. The first data module collects and processes the first detection data, and transmits the results of the first detection data to the gap preliminary analysis module;

[0017] Step 2: The initial gap analysis section processes the first detection data based on the machine learning model, identifies suspected gaps and records them;

[0018] Step 3: In the second data plate, the suspected position marked by the preliminary analysis plate of the gap is detected by applying the impact echo method to the suspected gap to obtain second detection data, and the second detection data is recorded and stored in the second data plate;

[0019] Step 4: Through the gap confirmation analysis panel, extract the second detection data to correct the first monitoring data, obtain the track gap detection data, and issue an early warning for the gap.

[0020] Compared with the prior art, the present invention has the following advantages:

[0021] 1. The present invention obtains the track delamination and crack status by collecting and analyzing the internal structure of the ballastless track. The radar detection module and the impact echo supplementary detection module, the data acquisition system and the data processing system can work separately or in collaboration.

[0022] 2. The first data block in the present invention is used to process the first monitoring data obtained by detecting the ballastless track through the radar monitoring module, making full use of the excellent performance of the ground penetrating radar in quickly detecting the ballastless track and shortening the construction period.

[0023] 3. The present invention can process ground penetrating radar data by training the support vector machine to obtain the suspected gap position. The machine learning model can be used to quickly judge the suspected gap position, reducing labor costs.

[0024] 4. According to the suspected crack position determined by the ground penetrating radar system, the present invention uses the support vector machine to send a signal, performs a supplementary impact echo test, obtains the second data block data, and corrects the first data block data. The impact echo method fully utilizes the characteristic of accurately detecting the gap to correct the first block data, making up for the shortcoming of insufficient accuracy of the first block, while avoiding the problem of the impact echo method requiring the arrangement of a large number of measuring points.

[0025] 5. The present invention has excellent performance in the field of roadbed engineering disease detection, with high detection efficiency, accurate results, and significant application effects. It has broad application prospects and commercial value. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is a schematic diagram of the structure of a track gap detection device;

[0027] Figure 2 This is a flow chart of a method for detecting gaps and cracks in ballastless track;

[0028] Figure 3 This is a schematic diagram of the data flow for the first detection of gaps in ballastless track;

[0029] Figure 4 This is a schematic diagram of the data flow for the second detection of gaps in ballastless track;

[0030] Figure 5 This is the principle diagram of shock echo detection of gaps and cracks;

[0031] Figure 6 Schematic diagram of the terminal device. DETAILED DESCRIPTION

[0032] The technical solution of the present invention is further described below with reference to the accompanying drawings, but is not limited thereto. Any modification or equivalent replacement of the technical solution of the present invention that does not depart from the spirit and scope of the technical solution of the present invention should be included in the scope of protection of the present invention.

[0033] The present invention provides a track gap detection device, such as Figure 1 As shown, the device includes a data acquisition system, a data processing system and a transmission system, wherein the data acquisition system is connected to the transmission system, and the transmission system is connected to the data processing system, wherein:

[0034] The data acquisition system is used to collect internal information of the ballastless track and includes a positioning module, a radar detection module, and an impact echo supplementary detection module. The positioning module is used to provide position information to the radar detection module; the radar detection module is used to initially collect internal information of the ballastless track; and the impact echo supplementary detection module is used to perform secondary detection on suspected ballastless track crack points to determine detailed information about the cracks.

[0035] The transmission system is used for information transmission between the data acquisition system and the data processing system, and can adopt one of USB interface transmission, 5G, 4G transmission or wireless transmission;

[0036] The data processing system is used to process the internal information of the ballastless track collected by the data acquisition system and obtain processing information, including a first data plate, a preliminary crack analysis plate, a second data plate, and a crack confirmation analysis plate, wherein: the first data plate is used to collect and process the ballastless track track plate data obtained by preliminary detection of the ground penetrating radar; the preliminary crack analysis plate judges the first data plate by means of machine learning, and is equipped with a positioning device to mark the location of suspected cracks; the second data plate is used to apply the impact echo detection method to the location of the suspected crack determined by the preliminary crack analysis plate, re-detect the marked location, and transmit the detection data to the crack confirmation analysis plate; the crack confirmation analysis plate is used to process and confirm a series of indicators of crack location, width, and shape.

[0037] The present invention provides a method for detecting track gaps. Figure 2 As shown, the method includes the following steps:

[0038] Step 1: Based on the ground penetrating radar method, the track is tested for gaps and cracks, and the first test data is collected and processed by the first data module, and the results of the first test data are transmitted to the gap preliminary analysis module. Figure 3 The specific steps are as follows:

[0039] Step 11: Convert the first detection data to obtain position interval data.

[0040] In this step, the main characteristic of ground-penetrating radar is its hyperbolic nature. Due to the extremely strict design and construction requirements for the track slabs in ballastless track structures, the distribution of rebar is spatially symmetrical, resulting in a spatially periodic echo signal. However, defects in the ballastless track are randomly distributed, resulting in a spatially non-periodic echo signal, representing a singular point. Therefore, the present invention can employ ground-penetrating radar to inspect ballastless track. However, ground-penetrating radar also has drawbacks: poor anti-interference capabilities and numerous constraints. During transmission, electromagnetic waves are subject to strong reflections from the rebar within the ballastless track. Therefore, ground-penetrating radar alone cannot accurately detect defects in railway sub-track structures. Therefore, it is necessary to perform an fk transform and an inverse fast Fourier transform on the first detection data to obtain position interval data.

[0041] In this step, the fk transformation formula is as follows:

[0042]

[0043] Where g(x, t) is the GPR value, frequency f is the number of GPR wave cycles per unit time (1 second), and is used to represent the Fourier transform of the GPR value in time. Wave number k is the number of wavelengths per unit distance (1 meter), and is used to represent the Fourier transform of the GPR value in space.

[0044] Ground Penetrating Radar Value: The radar waveform is divided into multiple curves, each of which represents a time function g(x i ,t j ), x i represents the i-th scan, t j represents the jth sampling point in the time history curve of the i-th scan. The multiple scans are combined to form the ground penetrating radar value g(x,t), that is:

[0045]

[0046] Among them, N is the number of radar channels and M is the number of collection points.

[0047] The value of the wave velocity v can be determined based on the relationship between period and frequency, and wave number and wavelength:

[0048]

[0049] Where T is the period of the GPR wave, λ is the wavelength of the GPR wave, and v is the velocity of the GPR wave.

[0050] In this step, after the first detection data undergoes the first transformation (i.e., the fk transform), the DC value of the first detection data is not located at the center of the matrix. Therefore, the first detection data can be offset. The parameter range data undergoes the second transformation to obtain position interval data. In this step, the parameter range data can be subjected to an inverse fast Fourier transform to obtain the position interval data for the ballastless track.

[0051] Step 12: Extract the first detection data feature based on the position interval data.

[0052] In this step, the position interval data is averaged and the averaged position interval data is used as the feature of the first detection data for extraction. The averaged position interval data is: the average value of all data points in the position interval data is calculated.

[0053] Step 2: The preliminary analysis section of the gap processes the first detection data based on the machine learning model, determines the suspected gap and records it.

[0054] In this step, the first detection data features need to be distinguished. The present invention uses a support vector machine to form a machine learning model, trains the model, and enables the model to automatically identify gaps and cracks and calibrate the locations of suspected gaps and cracks.

[0055] Step 3: Second data plate: The suspected position marked by the preliminary analysis plate of the gap is detected by applying the impact echo method to the suspected gap to obtain second detection data, and the second detection data is recorded and stored in the second data plate.

[0056] In this step, the impact echo method is performed on the calibration position to collect the echo signal data during the propagation of the impact echo on the ballastless track to obtain the second detection data. The characteristics of the echo signal data are extracted and quantified as one of the correction features for the gap detection. The following formula is used to solve it:

[0057]

[0058] Where: T is the thickness of the ballastless track slab, in mm; v c is the propagation velocity of longitudinal waves in concrete structures, measured in m / s; f is the frequency of the defect depth or slab thickness, corresponding to the peak frequency in the spectrum, measured in Hz; and β is the structural shape factor. For concrete ballastless track structures, β can range from 0.87 to 0.96.

[0059] According to the above formula, the data characteristics of the impact echo method can be sorted out by processing the data in a similar way as the first detection data. Figure 5 As shown in the schematic diagram, t1 and t2 correspond to the time when the peak value appears, and f1 and f2 correspond to the peak frequency.

[0060] Step 4: Through the gap confirmation analysis panel, extract the second detection data to correct the first monitoring data, obtain the track gap detection data, and issue an early warning for the gap.

[0061] In this step, the characteristics of the first detection data result are corrected based on the characteristics of the second detection data result to obtain a more accurate gap detection signal.

[0062] The present invention also provides a terminal processing device, such as Figure 6 As shown, it includes a memory, a processor, an input and output device, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned ballastless track gap detection method are implemented, wherein:

[0063] The processor, memory, input device, and output device communicate with each other via a communication bus. The memory is primarily used to store computer programs, which include program instructions. The processor is used to execute the program instructions stored in the memory. The processor is configured to invoke the program instructions to perform the functions of the modules / units in the above-described apparatus embodiments.

[0064] The processor may be a central processing unit, a digital signal processor, an application specific integrated circuit, etc.

[0065] The input device may include a camera, a microphone, a signal acquisition device, etc., which is connected to the device through a universal interface.

[0066] The output devices may include a display, speakers, etc.

[0067] The memory may include a read-only memory and a random access memory, and provides instructions and data to the processor.

Claims

1. A track gap detection device, characterized in that The device includes a data acquisition system, a data processing system and a transmission system, wherein the data acquisition system is connected to the transmission system, and the transmission system is connected to the data processing system, wherein: The data acquisition system includes a positioning module, a radar detection module, and an impact echo supplementary detection module; The positioning module is used to provide position information for the radar detection module; The radar detection module is used to preliminarily collect internal information of the ballastless track; The impact echo supplementary detection module is used to perform secondary detection on the suspected ballastless track points to determine the detailed information of the cracks; The data processing system includes a first data plate, a gap preliminary analysis plate, a second data plate, and a gap confirmation analysis plate; The first data block is used to collect and process ballastless track slab data obtained through preliminary detection by ground penetrating radar; The crack preliminary analysis panel uses machine learning to determine the first data panel and is equipped with a positioning device to mark the location of suspected cracks. The second data block is used to apply the impact echo detection method to the suspected crack position determined by the crack preliminary analysis block, re-detect the marked position, and transmit the detection data to the crack confirmation analysis block; The crack confirmation and analysis section is used to process and confirm a series of indicators of crack location, width and shape.

2. The track gap detection device according to claim 1 is characterized in that The transmission system adopts one of USB interface transmission, 5G, 4G transmission or wireless transmission.

3. A method for detecting track gaps using the device according to any one of claims 1-2, characterized in that The method comprises the following steps: Step 1: Using a ground penetrating radar method, a gap detection is performed on the non-operated track to obtain first detection data. The first data module collects and processes the first detection data, and transmits the results of the first detection data to the gap preliminary analysis module; Step 2: The initial gap analysis section processes the first detection data based on the machine learning model, identifies suspected gaps and records them; Step 3: In the second data plate, the suspected position marked by the preliminary analysis plate of the gap is detected by applying the impact echo method to the suspected gap to obtain second detection data, and the second detection data is recorded and stored in the second data plate; Step 4: Through the gap confirmation analysis panel, extract the second detection data to correct the first monitoring data, obtain the track gap detection data, and issue an early warning for the gap.

4. The method for detecting track gaps according to claim 3 is characterized in that The specific steps of step one are as follows: Step 11: convert the first detection data to obtain position interval data; Step 12: Extract the first detection data feature based on the position interval data.

5. The method for detecting track gaps according to claim 4 is characterized in that In the step 11, the first detection data is subjected to fk transformation and inverse fast Fourier transformation to obtain position interval data.

6. The method for detecting track gaps according to claim 5, characterized in that The fk transformation formula is as follows: Where g(x, t) is the GPR value, frequency f is the number of cycles of GPR wave propagation per unit time (1s), and is used to represent the Fourier transform of the GPR value in time. Wave number k is the number of wavelengths of GPR wave propagation per unit distance, and is used to represent the Fourier transform of the GPR value in space.

7. The method for detecting track gaps according to claim 6, characterized in that The calculation formula of the ground penetrating radar value g(x, t) is as follows: Among them, N is the number of radar channels and M is the number of collection points.

8. The method for detecting track gaps according to claim 3 is characterized in that In the step 2, a support vector machine is used to form a machine learning model, and the model is trained so that the model can automatically identify the gap area and calibrate the position of the suspected gap area.

9. The method for detecting track gaps according to claim 3, characterized in that In step 3, the impact echo method is performed on the calibration position to collect echo signal data during the propagation of the impact echo on the ballastless track to obtain second detection data. The characteristics of the echo signal data are extracted and quantified as one of the correction features for the gap detection. Specifically, the following formula is used to solve it: Where: T is the thickness of the ballastless track slab; v c is the propagation velocity of longitudinal waves in concrete structures; f is the frequency of defect depth or plate structure thickness, corresponding to the peak frequency in the spectrum; β is the structural shape coefficient.

10. A terminal processing device comprising a memory, a processor, an input / output device, and a computer program stored in the memory and executable on the processor, characterized in that When the processor executes the computer program, the steps of the track gap detection method according to any one of claims 3 to 9 are implemented.

Images