Method and device for detecting empty hoisting state of sleeper, electronic equipment, readable storage medium and program product

Abstract

The invention relates to an empty hoisting state detection method and device of a sleeper, electronic equipment, a computer readable storage medium and a computer program product. The method comprises the steps of obtaining a railway detection signal; the railway detection signal comprises a detection signal acquired on the railway track based on a ground penetrating radar carried on the railway vehicle when the railway vehicle runs on the railway track; time domain waveform difference information between the sleeper signal and the ballast bed signal at the detection position is determined according to the railway detection signal; the sleeper signal comprises an electromagnetic wave signal reflected by the sleeper; the ballast bed signal comprises an electromagnetic wave signal reflected by a ballast bed; according to the time domain waveform difference information, the distance between the sleeper at the detection position and the ballast bed is determined; and determining an empty hoisting detection result of the sleeper at the detection position according to the distance. By adopting the method, the accuracy of the empty hoisting state detection result of the sleeper can be improved.

Description

Technical Field

[0001] This application relates to the field of railway technology, and in particular to a method, apparatus, electronic device, computer-readable storage medium, and computer program product for detecting the unloaded state of railway sleepers. Background Technology

[0002] Ballasted track is one of the main structural forms of railway lines, consisting of rails, sleepers, ballast layers, and roadbed. Sleepers (also called sleepers) play a crucial role in evenly distributing the train load to the track bed. Sleeper detachment (i.e., a gap or insufficient support between the bottom of the sleeper and the track bed) is a common track defect that often leads to traffic safety issues.

[0003] The existing methods for detecting empty sleepers in related technologies mostly rely on the interpretation of overall track bed images, indirect indicators (such as energy loss or changes in scattering patterns), or the combination of dynamic data from track inspection vehicles. These methods lack sufficient accuracy and automation, and are greatly affected by factors such as ballast contamination, moisture content, and sleeper interference, making it difficult to accurately detect the empty sleeper status.

[0004] Therefore, there is a problem with the inaccuracy of sleeper empty-sleeve condition detection results in related technologies. Summary of the Invention

[0005] Therefore, it is necessary to provide a method, apparatus, electronic device, computer-readable storage medium, and computer program product for detecting the empty state of railway sleepers, which can improve the accuracy of the detection results of the empty state of railway sleepers, in order to address the above-mentioned technical problems.

[0006] Firstly, this application provides a method for detecting the unloaded state of railway sleepers, including:

[0007] Acquire railway detection signals; the railway detection signals include detection signals collected by ground-penetrating radar mounted on the railway vehicle when the railway vehicle is running on the railway track;

[0008] The time-domain waveform difference information between the sleeper signal and the ballast signal at the detection location is determined based on the railway detection signal; the sleeper signal includes the electromagnetic wave signal reflected by the sleeper; the ballast signal includes the electromagnetic wave signal reflected by the ballast.

[0009] Based on the time-domain waveform difference information, the distance between the sleeper and the track bed at the detection location is determined;

[0010] Based on the distance, the empty-sleeper detection result at the detection location is determined.

[0011] In one embodiment, determining the time-domain waveform difference information between the sleeper signal and the ballast signal at the detection location based on the railway detection signal includes:

[0012] Extract the sleeper signal and the track bed signal at the detection location from the railway detection signal;

[0013] Based on the track bed signal and the sleeper signal, a first time and a second time are determined respectively; the first time includes the time when the ground penetrating radar receives the electromagnetic wave signal reflected from the top of the track bed at the detection position; the second time includes the time when the ground penetrating radar receives the electromagnetic wave signal reflected from the top of the sleeper at the detection position.

[0014] The time-domain waveform difference information is generated based on the difference between the first time and the second time.

[0015] In one embodiment, determining the first time and the second time based on the track bed signal and the sleeper signal respectively includes:

[0016] Determine the signal peak values ​​of the sleeper signal and the track bed signal respectively;

[0017] The sampling time corresponding to the peak value of the track bed signal is taken as the first time.

[0018] The sampling time corresponding to the peak value of the sleeper signal is taken as the second time.

[0019] In one embodiment, determining the signal peak values ​​of the sleeper signal and the track bed signal respectively includes:

[0020] For any one of the sleeper signal and the track bed signal, the signal peak value of any one signal is searched using the Ricker wavelet template method or the time-domain waveform method;

[0021] Alternatively, the energy peak value in the time-frequency domain of any signal can be extracted using the continuous wavelet transform method, and used as the signal peak value of any signal.

[0022] In one embodiment, the distance is the distance between the top of the sleeper and the top of the track bed, and determining the empty sleeper detection result at the detection location based on the distance includes:

[0023] Obtain a preset distance threshold; the preset distance threshold is related to the thickness of the sleeper;

[0024] Based on whether the distance meets the preset distance threshold, it is determined whether the sleeper is in an empty state during the empty hoisting detection.

[0025] In one embodiment, after the step of acquiring the railway detection signal, the method further includes:

[0026] The railway detection signal is subjected to zero-bias removal processing to obtain the zero-bias removed railway detection signal;

[0027] The zero-bias-free railway detection signal is subjected to zero-point adjustment processing to obtain the zero-point adjusted railway detection signal;

[0028] The zero-point adjusted railway detection signal is subjected to phase shift migration processing in the frequency-wavenumber domain to obtain the migrated railway detection signal.

[0029] Secondly, this application also provides a sleeper unloaded status detection device, comprising:

[0030] The signal acquisition module is used to acquire railway detection signals; the railway detection signals include detection signals collected by ground-penetrating radar mounted on the railway vehicle when the railway vehicle is running on the railway track.

[0031] The information determination module is used to determine the time-domain waveform difference information between the sleeper signal and the ballast signal at the detection location based on the railway detection signal; the sleeper signal includes the electromagnetic wave signal reflected by the sleeper; the ballast signal includes the electromagnetic wave signal reflected by the ballast.

[0032] The distance determination module is used to determine the distance between the sleeper and the track bed at the detection location based on the time-domain waveform difference information.

[0033] The result determination module is used to determine the empty-sleeper detection result at the detection location based on the distance.

[0034] Thirdly, this application also provides an electronic device. The electronic device includes a memory and a processor, the memory storing a computer program that, when executed by the processor, implements the steps of the method described above.

[0035] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of the above-described method.

[0036] Fifthly, this application also provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps of the above-described method.

[0037] The aforementioned method, apparatus, electronic equipment, computer-readable storage medium, and computer program product for detecting the empty-sleeve status of railway sleepers acquire railway detection signals; these railway detection signals include detection signals collected from the railway track by ground-penetrating radar mounted on the railway vehicle when the vehicle is running on the track; based on the railway detection signals, the time-domain waveform difference information between the sleeper signal and the ballast signal at the detection location is determined; the sleeper signal includes electromagnetic wave signals reflected by the sleeper; the ballast signal includes electromagnetic wave signals reflected by the ballast; based on the time-domain waveform difference information, the distance between the sleeper and the ballast at the detection location is determined; and based on the distance, the empty-sleeve detection result of the sleeper at the detection location is determined.

[0038] Thus, by collecting railway detection signals from ground-penetrating radar mounted on railway vehicles, the time-domain waveform difference information between the sleeper signal and the ballast signal at the detection location can be determined. The sleeper signal includes the electromagnetic wave signal reflected by the sleeper; the ballast signal includes the electromagnetic wave signal reflected by the ballast. Therefore, the distance between the sleeper and the ballast at the detection location can be accurately identified based on the time-domain waveform difference information. Based on this distance, the empty sleeper detection result at the detection location can be determined. This enables the rapid and accurate detection of the sleeper's empty status by utilizing the time-domain waveform difference between the ground-penetrating radar signal above the sleeper and the ballast surface, as well as the physical characteristics of the empty sleeper, effectively improving the accuracy of the sleeper empty status detection results. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 This is a flowchart illustrating a method for detecting the unloaded state of a railway sleeper in one embodiment.

[0041] Figure 2 This is a flowchart illustrating another method for detecting the unloaded state of a railway sleeper in one embodiment;

[0042] Figure 3 This is a schematic diagram of a ground-penetrating radar detecting a railway in one embodiment;

[0043] Figure 4 This is a schematic diagram of a GPR signal after zero-point adjustment in one embodiment;

[0044] Figure 5 This is a schematic diagram of a peak search based on Ricker wavelet template matching in one embodiment;

[0045] Figure 6 This is a schematic diagram of the identification process for a sleeper area in one embodiment;

[0046] Figure 7 This is a schematic diagram of a simulation model of a ballast railway track bed suspended without load in one embodiment;

[0047] Figure 8 This is a schematic diagram of a process for determining whether a sleeper is being hoisted empty, as described in one embodiment.

[0048] Figure 9 This is a schematic diagram of a simulation result in one embodiment;

[0049] Figure 10 This is a comparison diagram of the signal and peak time of a sleeper and the top of the track bed in one embodiment;

[0050] Figure 11(a) is a schematic diagram of the peak search results for a type of sleeper in one embodiment;

[0051] Figure 11(b) is a schematic diagram of the peak search results for a normal track bed in one embodiment;

[0052] Figure 11(c) is a schematic diagram of peak search results for a type of empty area track bed in one embodiment;

[0053] Figure 11(d) is a schematic diagram of the peak search results for another type of empty area track bed in one embodiment;

[0054] Figure 11(e) is a schematic diagram of the peak search results for another type of empty area track bed in one embodiment;

[0055] Figure 11(f) is a schematic diagram of the peak search results for a suspended track bed in another embodiment;

[0056] Figure 12 This is a schematic diagram of a measured railway signal in one embodiment;

[0057] Figure 13 This is a schematic diagram showing a signal comparison between a sleeper and a track bed based on measured data in one embodiment.

[0058] Figure 14(a) is a schematic diagram of peak search results for a sleeper in another embodiment;

[0059] Figure 14(b) is a schematic diagram of peak search results for a type of track bed in another embodiment;

[0060] Figure 15(a) is a photograph taken at a railway site in one embodiment;

[0061] Figure 15(b) shows another railway scene photograph in one embodiment;

[0062] Figure 16 This is a structural block diagram of a sleeper empty-sleeve status detection device in one embodiment;

[0063] Figure 17 This is a diagram of the internal structure of an electronic device in one embodiment. Detailed Implementation

[0064] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0065] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0066] In one embodiment, such as Figure 1 As shown, a method for detecting the empty state of railway sleepers is provided. This embodiment illustrates the application of this method to an electronic device. It is understood that the electronic device can be a terminal, a server, or a system including both a terminal and a server, and the detection is achieved through interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0067] Step S110: Obtain railway detection signals.

[0068] Among them, railway detection signals include detection signals collected by ground-penetrating radar mounted on railway vehicles when they are running on railway tracks.

[0069] In practical applications, Ground Penetrating Radar (GPR), as a high-frequency electromagnetic wave non-destructive testing technology, has been widely used in areas such as railway ballast bed contamination assessment, subgrade defect detection, and interlayer void identification in ballastless track. GPR transmits electromagnetic waves towards the target via a transmitting antenna. Due to differences in materials, densities, and levels of contamination within the target, the electromagnetic waves are reflected at interfaces where electromagnetic properties change. GPR receives the reflected echo signals through a high-sensitivity receiving unit and interprets information such as target depth, medium structure, and properties based on changes in time delay, waveform, and spectral characteristics. By mounting the GPR detection antenna on railway vehicles, non-stop testing can be achieved, improving detection efficiency and effectively ensuring safe railway operation.

[0070] Among them, railway detection signals refer to signals used to detect the unloaded status of sleepers on railway tracks.

[0071] Ground-penetrating radar (GPR) transmits electromagnetic signals to the railway track via a transmitting antenna (TX) and receives the reflected electromagnetic signals via a receiving antenna (RX). Electronic equipment processes the reflected electromagnetic signals from the railway track obtained by GPR to acquire railway detection signals.

[0072] In some embodiments, electronic devices can locate sleeper regions within railway detection signals. A sleeper region refers to a sampling point or time range of a sleeper. Based on the located sleeper region, the temporal waveform difference information between the sleeper signal and the ballast signal at the detection location can be further determined. In some embodiments, a deep learning convolutional neural network (CNN) can be used to locate the corresponding sleeper region within the railway detection signal. In other embodiments, the user can manually specify the sampling point or time range of the sleeper to determine the sleeper region. Thus, accurately identifying the sleeper region from the overall railway detection signal collected by onboard ground-penetrating radar during railway vehicle operation can effectively eliminate interference from irrelevant signals such as roadbed and rail clutter. This provides a foundation for subsequent steps such as peak extraction, distance calculation, and empty-sleeper status determination for sleepers and ballast, avoiding feature extraction deviations and judgment errors caused by the mixing of irrelevant signals, reducing false positives and false negatives in empty-sleeper detection, and reducing unnecessary computation.

[0073] Step S120: Determine the time-domain waveform difference information between the sleeper signal and the track bed signal at the detection location based on the railway detection signal.

[0074] Among them, the sleeper signal includes the electromagnetic wave signal from the top of the sleeper.

[0075] Among them, the track bed signal includes the electromagnetic wave signal from the top of the track bed.

[0076] The detection location is the location that includes the sleepers.

[0077] In some embodiments, the electronic device can determine the time-domain waveform difference information between the sleeper signal and the ballast signal at the detection location based on the railway detection signal. Specifically, the time-domain waveform difference information can be the difference between the sleeper signal and the ballast signal in the time domain.

[0078] Step S130: Determine the distance between the sleeper and the track bed at the detection location based on the time-domain waveform difference information.

[0079] In practice, the electronic device can determine the distance between the sleeper and the track bed at the detection location based on the time-domain waveform difference information. Specifically, it can determine the distance between the top or bottom of the sleeper and the top of the track bed at the detection location.

[0080] In some embodiments, the time-domain difference information between the sleeper signal and the track bed signal can specifically be the difference between the time when the ground penetrating radar receives the electromagnetic wave signal reflected from the top of the track bed at the detection location and the time when the ground penetrating radar receives the electromagnetic wave signal reflected from the top of the sleeper at the detection location. Thus, the distance between the top of the sleeper and the top of the track bed at the detection location can be determined based on the time-domain waveform difference information.

[0081] Step S140: Determine the empty-sleeper inspection result at the inspection location based on the distance.

[0082] In practice, the electronic device can determine whether the sleeper at the detection location is in an unloaded state based on the distance between the sleeper and the track bed at the detection location, and output the unloaded sleeper detection result.

[0083] In the above-mentioned method for detecting the empty state of railway sleepers, railway detection signals are acquired. These signals include those collected by ground-penetrating radar mounted on railway vehicles when they are running on the tracks. The time-domain waveform difference information between the sleeper signal and the ballast signal at the detection location is determined based on the railway detection signals. The sleeper signal includes electromagnetic wave signals reflected by the sleeper. The ballast signal includes electromagnetic wave signals reflected by the ballast. The distance between the sleeper and the ballast at the detection location is determined based on the time-domain waveform difference information. The empty-sleeve detection result of the sleeper at the detection location is determined based on the distance.

[0084] Thus, by collecting railway detection signals from ground-penetrating radar mounted on railway vehicles, the time-domain waveform difference information between the sleeper signal and the ballast signal at the detection location can be determined. The sleeper signal includes the electromagnetic wave signal reflected by the sleeper; the ballast signal includes the electromagnetic wave signal reflected by the ballast. Therefore, the distance between the sleeper and the ballast at the detection location can be accurately identified based on the time-domain waveform difference information. Based on this distance, the empty sleeper detection result at the detection location can be determined. This enables the rapid and accurate detection of the sleeper's empty status by utilizing the time-domain waveform difference between the ground-penetrating radar signal above the sleeper and the ballast surface, as well as the physical characteristics of the empty sleeper, effectively improving the accuracy of the sleeper empty status detection results.

[0085] In some embodiments, determining the time-domain waveform difference information between the sleeper signal and the ballast signal at the detection location based on the railway detection signal includes: extracting the sleeper signal and the ballast signal at the detection location from the railway detection signal; determining a first time and a second time based on the ballast signal and the sleeper signal, respectively; the first time includes the time when the ground penetrating radar receives the electromagnetic wave signal reflected from the top of the ballast at the detection location; the second time includes the time when the ground penetrating radar receives the electromagnetic wave signal reflected from the top of the sleeper at the detection location; and generating time-domain waveform difference information based on the difference between the first time and the second time.

[0086] In some embodiments, the first time can specifically be the two-way travel time from when the ground penetrating radar emits an electromagnetic wave signal at the detection location to when it receives the electromagnetic wave signal reflected from the top of the track bed.

[0087] In some embodiments, the second time can specifically be the two-way travel time from when the ground-penetrating radar emits an electromagnetic wave signal at the detection location to when it receives the electromagnetic wave signal reflected from the top of the sleeper.

[0088] In practical implementation, during the process of determining the time-domain waveform difference information between the sleeper signal and the ballast signal at the detection location based on the railway detection signal, the electronic equipment can extract the sleeper signal and the ballast signal at the detection location from the railway detection signal. Based on the ballast signal and the sleeper signal, a first time and a second time are determined respectively. Specifically, the first time can be the two-way travel time from when the ground-penetrating radar emits an electromagnetic wave signal at the detection location to when it receives the electromagnetic wave signal reflected from the top of the ballast. Similarly, the second time can be the two-way travel time from when the ground-penetrating radar emits an electromagnetic wave signal at the detection location to when it receives the electromagnetic wave signal reflected from the top of the sleeper. Thus, based on the difference between the first time and the second time, time-domain waveform difference information can be generated, which specifically characterizes the time-domain difference between the sleeper signal and the ballast signal.

[0089] The technical solution of this embodiment detects electromagnetic wave signals reflected from the sleeper area. It extracts the sleeper signal and ballast signal at the detection location from the railway detection signals. Based on the ballast signal and sleeper signal, a first time and a second time are determined. The first time includes the time when the ground-penetrating radar receives the electromagnetic wave signal reflected from the top of the ballast at the detection location; the second time includes the time when the ground-penetrating radar receives the electromagnetic wave signal reflected from the top of the sleeper at the detection location. Based on the difference between the first time and the second time, time-domain waveform difference information is generated. Thus, by accurately reflecting the time-domain waveform difference information between the sleeper signal and the ballast signal at the detection location based on the difference between the time when the ground-penetrating radar receives the electromagnetic wave signal reflected from the top of the ballast at the detection location and the time when the ground-penetrating radar receives the electromagnetic wave signal reflected from the top of the sleeper at the detection location, the distance between the two at the detection location can be calculated.

[0090] In some embodiments, the time-domain waveform difference information can specifically be the difference in the signal peak values ​​of the sleeper signal and the track bed signal in the time domain, so as to determine the distance between the top of the sleeper and the top of the track bed at the detection position based on the time difference from the top of the sleeper to the top of the track bed, thereby realizing the quantitative identification of sleeper slewing.

[0091] In some embodiments, determining a first time and a second time based on the track bed signal and the sleeper signal includes: determining the signal peak values ​​of the sleeper signal and the track bed signal respectively; taking the sampling time corresponding to the signal peak value of the track bed signal as the first time; and taking the sampling time corresponding to the signal peak value of the sleeper signal as the second time.

[0092] In the specific implementation, during the process of determining the first time and the second time based on the track bed signal and the sleeper signal, the electronic equipment can determine the signal peak values ​​of the sleeper signal and the track bed signal respectively; the sampling time corresponding to the signal peak value of the track bed signal is taken as the first time; and the sampling time corresponding to the signal peak value of the sleeper signal is taken as the second time.

[0093] In some embodiments, the sampling time corresponding to the signal peak can be determined based on the sampling point corresponding to the signal peak and the sampling frequency.

[0094] The technical solution of this embodiment determines the peak values ​​of the sleeper signal and the track bed signal separately; the sampling time corresponding to the peak value of the track bed signal is taken as the first time; and the sampling time corresponding to the peak value of the sleeper signal is taken as the second time. Thus, by using the sampling times corresponding to the peak values ​​of the track bed signal and the sleeper signal as the first and second times respectively, the time when the ground-penetrating radar receives the electromagnetic wave signals reflected from the top of the track bed and the top of the sleeper at the detection location can be accurately determined, providing a precise and reliable basis for subsequent distance calculation between the two and determination of the empty-sleeper status.

[0095] In some embodiments, determining the signal peak values ​​of the sleeper signal and the track bed signal respectively includes: searching for the signal peak value of any signal using the Ricker wavelet template method or the time-domain waveform method for any of the sleeper signals and track bed signals; or extracting the energy peak value in the time-frequency domain of any signal using the continuous wavelet transform method as the signal peak value of any signal.

[0096] In practice, when determining the signal peak values ​​of the sleeper signal and the track bed signal, the electronic equipment can search for the signal peak value of any signal using the Ricker wavelet template method or the time-domain waveform method; or, it can extract the energy peak value in the time-frequency domain of any signal using the continuous wavelet transform method, and use it as the signal peak value of any signal.

[0097] In the process of extracting the energy peak value of any signal in the time-frequency domain using the continuous wavelet transform method, the signal can be processed column by column of the column vector to obtain a matrix. The matrix is ​​then subjected to modulo operation to obtain energy data. The global maximum value is determined in the energy data, thereby determining the energy peak value.

[0098] Among them, time-domain methods (Ricker wavelet template method and time-domain waveform method) are simple and efficient in operation, and can quickly capture the time-domain peak characteristics of signals, making them suitable for the rapid detection scenario of vehicle-mounted GPR without stopping operation; the continuous wavelet transform method can accurately extract the time-frequency domain energy peak of signals, effectively avoid noise interference, and improve the accuracy of peak extraction under complex working conditions.

[0099] The technical solution of this embodiment searches for the signal peak of either the sleeper signal or the track bed signal using the Ricker wavelet template method or the time-domain waveform method; alternatively, it extracts the energy peak in the time-frequency domain of either signal using the continuous wavelet transform method, which is then used as the signal peak of that signal. This provides three flexible signal peak extraction methods—the Ricker wavelet template method, the time-domain waveform method, and the continuous wavelet transform method—that can be selected according to actual detection needs, offering strong adaptability. Furthermore, by extracting peak values ​​separately for both the sleeper and track bed signals, it provides accurate and reliable signal characteristic data for subsequent distance calculations and empty-sleeve status determination, reducing false detections and missed detections caused by peak value extraction deviations and ensuring the accuracy of sleeper empty-sleeve detection results.

[0100] In some embodiments, the distance is the distance between the top of the sleeper and the top of the track bed. Based on the distance, the empty sleeper detection result in the sleeper area is determined, including: obtaining a preset distance threshold; the preset distance threshold is associated with the thickness of the sleeper in the railway track; and determining whether the empty sleeper detection result is in an empty state based on whether the distance meets the preset distance threshold.

[0101] The preset distance threshold is related to the thickness of the railway sleepers. In practical applications, the preset distance threshold can be adjusted to take into account the sleeper type and the dielectric constant of the ballast bed; for example, the preset distance H0 = sleeper thickness + compensation factor. In some other embodiments, the preset distance threshold can also be equal to the sleeper thickness.

[0102] In specific implementation, the distance between the sleeper and the track bed at the detection location can be the distance between the top of the sleeper and the top of the track bed. When the electronic device determines the empty sleeper detection result at the detection location based on the distance, it can obtain a preset distance threshold related to the thickness of the sleeper in the railway track. Based on whether the distance meets the preset distance threshold, it determines whether the empty sleeper detection result is in an empty state.

[0103] For example, if the preset distance threshold is equal to the sleeper thickness, if the distance between the top of the sleeper and the top of the track bed at the detection location is greater than or equal to the preset distance threshold, the sleeper at the detection location is determined to be in an empty state. If the distance is less than the preset distance threshold, the sleeper at the detection location is determined to be not in an empty state.

[0104] The technical solution of this embodiment obtains a preset distance threshold; the preset distance threshold is related to the thickness of the railway sleepers; and the result of the empty-sleeping detection is determined based on whether the distance between the top of the sleeper and the top of the ballast bed is greater than the preset distance threshold. Thus, by linking the preset distance threshold to the sleeper thickness, the threshold setting is made to fit the actual structural characteristics of the railway track, improving the rationality and accuracy of the empty-sleeping determination; at the same time, using whether the distance between the top of the sleeper and the top of the ballast bed exceeds the threshold as the determination criterion simplifies the conversion logic from detection signal characteristics to empty-sleeping results, reduces the complexity of signal processing and discrimination, and improves the efficiency of detection operations.

[0105] For the ease of understanding of those skilled in the art, Figure 2 A flowchart illustrating another method for detecting the unloaded state of railway sleepers is provided, including the following steps:

[0106] Step 1: Collect data on railway tracks, such as ballasted railways, using ground-penetrating radar (GPR) to obtain a GPR dataset. That is, to acquire the railway detection signal (GPR signal), specifically including step S11:

[0107] Step S11, GPR performs on-site data collection: The GPR system is used to collect on-site data of the ballast railway.

[0108] In step S11, during the on-site data acquisition process, such as Figure 3 The diagram illustrates a GPR (Gas Resonance Process) method for railway inspection. Sampling parameters require a sampling interval of <10cm, an antenna frequency of ≥1.5GHz, an antenna suspension height of ≥10cm, and a sampling frequency 10-30 times the antenna's main frequency. Therefore, by optimizing the sampling parameters (antenna ≥1.5GHz, spacing <10cm) for the characteristics of unloaded sleepers, real-time onboard inspection is ensured.

[0109] Step 2: GPR signal post-processing, specifically including steps S21 to S23:

[0110] Step S21: Perform zero-bias removal processing on the railway detection signal (GPR signal) to obtain the zero-bias removal railway detection signal.

[0111] In the specific implementation, GPR data is collected in step one. At that time, due to the drift of the electronic device itself, the acquired signal has a signal offset, so it is necessary to remove the zero offset, as shown in equation (1):

[0112] (1)

[0113] In formula (1) This indicates the number of sampling points in each A-scan (amplitude scan). This indicates the number of channels acquired by GPR. Indicates the first One sampling point.

[0114] Step S22: Perform zero-point adjustment processing on the railway detection signal after zero offset processing to obtain the zero-point adjusted railway detection signal.

[0115] In practical implementation, because the GPR (Gas Track Relay) is suspended, the direct wave appears in the air layer. Taking a GPR antenna with a center frequency of 1.5 GHz as an example, with a wavelength λ1.5 = 20 cm, the GPR is excited in the form of Ricker wavelet, which will form the maximum peak value in the direct wave. This will affect the subsequent positioning of sleepers and empty sling areas, and the direct wave needs to be removed. Figure 4 The diagram illustrates a zero-point adjustment of a GPR signal. To avoid the influence of the Ricker wavelet, zero-point adjustment is performed by searching for the sampling point corresponding to the peak value of the first Ricker wavelet. The deviation will be further adjusted based on the sampling frequency and wavelength. The formula for zero-point adjustment is shown in formula (2):

[0116] (2)

[0117] In formula (2) The signal after zero bias removal Each A-scan signal in the middle reaches the sampling point corresponding to the extreme point of the wave; As the offset point, taking 1.5GHz as an example, set... This can avoid interference from direct waveside lobe signals.

[0118] In step S22, template matching is used to locate the peak value of the first Ricker wavelet. The search process is as follows: Figure 5 As shown, steps S221 to S223 are included:

[0119] Step S221: Generate Ricker wavelet.

[0120] (3)

[0121] In the formula The center frequency of the GPR antenna. For time; adjust formula (3) according to the actual sampling frequency at the site. Discretization is performed, and the sampling time is... The time of each wavelength.

[0122] Step S222: Cross-correlation calculation.

[0123] For the signal matrix of GPR post-processing The k-th signal is denoted as sk, which is the k-th column in the C matrix, and its time-domain expression is: The standard cross-correlation function is used for calculation. and The cross-correlation results are shown in formula (4):

[0124] (4)

[0125] In formula (4) For the cross-correlation results, the cross-correlation function can be called in the calculation software to implement the template Ricker wavelet and GPR post-processing signal. The calculation.

[0126] Step S223: Search for the maximum value to obtain the delay. Find the peak value of the Ricker wavelet.

[0127] Step S23: Perform phase shifting processing on the zero-point adjusted railway detection signal in the frequency-wavenumber domain to obtain the shifted railway detection signal.

[0128] In practice, the FK shift (also known as Stolt shift or ω-k shift) is a phase shift shift in the frequency-wavenumber domain, used to focus the diffraction hyperbola and correct tilt angle reflection. The calculation formula is as shown in formula (5):

[0129] (5)

[0130] The FK offset can be directly invoked in the program using the FK function in the calculation software, with the wave velocity set to v = 0.3 m / ns and the action time range specified as t = [0, 6] ns. After the FK offset, the GPR acquired signal becomes a matrix. .

[0131] Thus, in the signal post-processing, zero-point adjustment (offsetting the p0 point to avoid direct wavelet sidelobe interference) and FK offset focusing diffraction hyperbola are introduced to improve shallow resolution.

[0132] Step 3: Locating the sleeper area.

[0133] Method 1: Manually specify the sampling points or time range of the sleepers. Specify the sleeper area ns0=[m1,m2]. Based on the GPR antenna frequency and sampling points, it is recommended that ns0=[1,100].

[0134] Method 2: Locate the sleepers using a convolutional neural network (CNN) developed through deep learning, obtaining the sampling point range ns0=[m1,m2] for the sleeper area. For example... Figure 6 As shown, a schematic diagram of a sleeper area identification process is provided. A 640*640 image is exported using GPR post-processing software, the sleeper area is labeled, and a YOLOv11 model suitable for sleeper detection is trained using transfer learning. The GPR post-processed data... A 640*640 image is generated according to the 640 track intervals and imported into the YOLOv11 model to obtain the sleeper area range ns0=[m1,m2]. Here, m1 and m2 are obtained by linearly mapping the image size to the number of track sampling points.

[0135] Step 4: Determine the unloaded status of the sleepers, including steps S41 to S43:

[0136] Step S41: Search for the sampling time corresponding to the peak value of the signal.

[0137] Method 1: Utilizing the characteristics of the reflection amplitude from the sleepers and the top surface of the track bed, the Ricker wavelet template method is used to search for the peak value of the GPR signal. The post-processed GPR signal is then processed column-wise. Perform a column-by-column search for the signal peak A01. Representing the List, Alternatively, the peak value A01 can be directly searched within the sampling range of ns0=[m1,m2] in the sleeper region using the time-domain waveform method. The sampling point K0 corresponding to peak value A01 is found, and the corresponding sampling time ts=K0*dt is obtained, thereby determining the position of the sleeper or the top surface of the track bed.

[0138] Method 2: Post-process the GPR signal According to column vector The matrix is ​​obtained by performing CWT (Continuous Wavelet Transform) column by column. and the matrix The matrix is ​​obtained by taking the modulus. .in Represents a frequency array. Represents the array of time-domain sampling points. and These are the original CWT transformation result and the result after modulo operation, respectively.

[0139] Search Matrix The maximum value is obtained using the max function. The maximum energy value A02 and the corresponding column. .right No. Column vector Search for the row position corresponding to the maximum energy value A02 to obtain the row vector position. .according to and The time corresponding to the peak can be calculated separately. and peak frequency The peak frequency range can be used to constrain the maximum value range.

[0140] Thus, by combining time-domain waveform peak search (Ricker wavelet template matching) and continuous wavelet transform (CWT) time-frequency domain energy peak extraction, the time difference Δt and height H from the top of the sleeper to the track bed interface are calculated, enabling quantitative identification of empty sleepers. This dual-mode method is more robust than traditional amplitude attenuation or scattering indices, providing an interactive verification basis for judging the status of empty sleepers.

[0141] Step S42: Calculate the distance between the top of the track bed and the top of the sleeper.

[0142] The peak time at the top of the sleeper (the sampling time corresponding to the signal peak) is ts, and the peak time at the top of the track bed is t. The distance from the top of the track bed to the top of the sleeper is H = (t - ts) * v / 2, where v is the electromagnetic wave speed, which is v = 0.3 m / ns in air, and t is in ns. Therefore, if the influence of air humidity is not considered, the wave speed is not corrected, and the distance is calculated according to H = 0.15(t - ts); if the air humidity is too high, the wave speed v is calculated according to... calculate, Humid air.

[0143] Step S43: Set a preset distance threshold H0. If H≥H0, then determine that the sleeper is in an unloaded state.

[0144] In normal railways, the distance from the top of the sleeper to the top surface of the ballast layer is approximately 5cm to 10cm, and the thickness of the concrete sleeper in the center of the ballast bed is generally 18cm. Missing ballast can easily cause sleepers to be suspended in place; therefore, a preset distance threshold H0 = 18cm is set. When H ≥ H0, the top surface of the ballast bed is flush with or lower than the bottom of the sleeper, indicating that two adjacent sleepers are in a suspended state. By combining the track number and station number of the suspended area, the distribution of suspended sleepers can be determined, providing a basis for railway maintenance.

[0145] In summary, this application utilizes the time-domain and time-frequency domain characteristics of high-frequency GPR to achieve non-destructive and rapid identification of sleeper ballast gaps. The detection antenna can be mounted on the locomotive for non-stop inspection, improving detection efficiency and effectively ensuring safe railway operation. During the identification of ballast gaps, by combining the spectral characteristics of GPR and the physical characteristics of ballast gaps, and considering the presence of ballast defects in the track bed when a ballast gap occurs, using the top of the sleeper as a reference height and the time difference Δt between the track bed interface and the top of the sleeper as the feedback quantity, a threshold is set to quickly and accurately determine the sleeper's ballast gap status. Furthermore, using the sleeper thickness (18cm) as a reference, setting H≥H0 to determine ballast gaps, and combining continuous track anomalies to determine the location and length, provides a basis for maintenance decisions. Moreover, this method does not require dynamic loading and is suitable for non-stop inspection.

[0146] To verify the effectiveness of the method in this application, such as Figure 7 As shown, a simulation model of a ballasted railway track bed with open sections was established. In the model, the sleeper thickness Hs = 18cm, the sleeper spacing is 60cm, and the height of the track surface from the sleeper in a normal track bed is 9cm. Four open sections were set up, with section S1 being the critical open section, where the distance from the top of the track surface to the bottom of the sleeper is H1 = 0cm; and sections S2 to S4 being the three open sections, with distances from the top of the track bed to the bottom of the sleeper of 4, 8, and 12cm, respectively. Simulations were performed using simulation software with an antenna frequency of 1500MHz, a track spacing of 5cm, a dielectric constant of 1 for the open sections and air, a dielectric constant of 9 for the sleepers, a dielectric constant of 4 for the track bed, and a dielectric constant of 12 for the roadbed. The original GPR data matrix was obtained after the simulation. The matrix is ​​obtained after post-processing analysis. ,refer to Figure 8 Determine the unloaded status of the sleepers and extract the data matrix. Perform peak search column by column according to steps three and four above, calculate the relative distance H from the top of the track bed to the top of the sleeper. If H ≥ H0 (H0 = 18cm), then it is judged as an empty hoist.

[0147] The B-scan corresponding to the simulation model is as follows: Figure 9 ,from Figure 9 The sleepers, the normal track bed area S0, and the unsupported areas S1~S4 can be effectively observed. The top of the track bed in the unsupported areas is lower than the interface of the normal track bed. A-scan signals are extracted from the top of the sleepers, the top of the track bed in the normal track bed area S0, and the top of the track bed in the unsupported areas S1~S4.

[0148] Using the time-domain search method, the corresponding peak times are obtained, and these A-scan signals and peak times are placed in a... Figure 10The diagram shows the sampling times for the top of the sleeper in the normal track bed area S0, and ts0 for the top of the track bed in the normal track bed area. ts1 to ts4 represent the sampling times for the top of the track bed in the empty track bed areas S1 to S4, respectively. The distance from the normal track surface to the top of the sleeper can be calculated as HS0 = 9.38 cm, which is 0.38 cm different from the simulation model's set height of 9 cm. This error is due to electromagnetic wave resolution and is generally less than 0.5 cm. Similarly, the distances from the top of the track bed to the top of the sleeper in the four empty track bed areas are obtained as follows: HS1 = 18.46 cm, HS2 = 22.57 cm, HS3 = 26.67 cm, and HS4 = 37.22 cm, all greater than 18 cm, indicating that these areas are in an empty track bed state.

[0149] The peak values ​​were searched using the time-frequency domain CWT method, and the CWT results for sleepers, normal track bed, and unsupported track bed are shown in Figure 11(a), (b), (c), (d), (e), and (f). The upper part of the figures displays the searched peak time, frequency, and energy value. "+" in the figures indicates the peak position, and two decimal places are retained. The peak times for the top of the sleeper, the top of the normal track bed, and the top of the unsupported track bed are ts=1.58ns, ts0=2.17ns, ts1=2.81ns, ts2=3.03ns, ts3=3.28ns, and ts4=4.02ns, respectively. The original CWT energy with peak times different from the time-domain waveform peaks exhibits dispersion; the peaks correspond to interface points. Based on the sleeper top ts=1.58ns, the distances H between the top of the normal track bed and the top of the sleeper in the unloaded areas 1-4 are calculated to be 8.85cm, 18.45cm, 21.75cm, 25.5cm and 36.6cm respectively. It can be seen that the distance H between the top of the track bed and the top of the sleeper in the four unloaded areas is greater than H0, indicating that the sleeper is in an unloaded state.

[0150] To further verify the generality of the method, analysis was performed on measured GPR data, and the processing method was consistent with the simulation results. The test data, after GPR post-processing, was obtained as follows: Figure 12 The B-scan diagram shown extracts and compares sleeper and track bed signals near 1420m. The results are as follows: Figure 13 As shown, the peak times were 0.8965ns and 2.125ns, respectively. The calculated distance H from the top surface of the track bed to the top surface of the sleeper at 1420m was 18.43cm, indicating that the sleeper at 1420m was in an unsupported state.

[0151] Similarly, the peak time was searched using the CWT method, and the calculation results are shown in Figure 14(a) and Figure 14(b). The peak times for the sleeper and the track bed are 0.73ns and 2.09ns, respectively. The calculated distance from the top of the track bed to the top of the sleeper is H=20.04cm, indicating that the sleeper in this area is in an unloaded state.

[0152] Figure 15(a) is a photo taken on site, and Figure 15(b) is a magnified view of a portion of Figure 15(a), which verifies the correctness of the judgment result of the method. If the sleepers from 1416m to 1422m are further judged, it can be determined that the sleepers in this area are in an unloaded state. The method of this application provides a scientific method for non-destructive testing and automatic judgment of sleepers.

[0153] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0154] Based on the same inventive concept, this application also provides a sleeper empty-sleeve state detection device for implementing the sleeper empty-sleeve state detection method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations of one or more sleeper empty-sleeve state detection device embodiments provided below can be found in the limitations of the sleeper empty-sleeve state detection method described above, and will not be repeated here.

[0155] In one exemplary embodiment, such as Figure 16 As shown, a sleeper empty-sleeping state detection device is provided, including: a signal acquisition module 1610, an information determination module 1620, a distance determination module 1630, and a result determination module 1640, wherein:

[0156] The signal acquisition module 1610 is used to acquire railway detection signals; the railway detection signals include detection signals collected by ground-penetrating radar mounted on the railway vehicle when the railway vehicle is running on the railway track.

[0157] The information determination module 1620 is used to determine the time-domain waveform difference information between the sleeper signal and the ballast signal at the detection location based on the railway detection signal; the sleeper signal includes the electromagnetic wave signal reflected by the sleeper; the ballast signal includes the electromagnetic wave signal reflected by the ballast.

[0158] The distance determination module 1630 is used to determine the distance between the sleeper and the track bed at the detection position based on the time-domain waveform difference information.

[0159] The result determination module 1640 is used to determine the empty-sleeper detection result at the detection position based on the distance.

[0160] In one embodiment, the information determination module 1620 is specifically used to extract the sleeper signal and the track bed signal at the detection location from the railway detection signal; determine a first time and a second time based on the track bed signal and the sleeper signal; the first time includes the time when the ground penetrating radar receives the electromagnetic wave signal reflected from the top of the track bed at the detection location; the second time includes the time when the ground penetrating radar receives the electromagnetic wave signal reflected from the top of the sleeper at the detection location; and generate the time-domain waveform difference information based on the difference between the first time and the second time.

[0161] In one embodiment, the information determination module 1620 is specifically used to determine the signal peak values ​​of the sleeper signal and the track bed signal respectively; to take the sampling time corresponding to the signal peak value of the track bed signal as the first time; and to take the sampling time corresponding to the signal peak value of the sleeper signal as the second time.

[0162] In one embodiment, the information determination module 1620 is specifically used to search for the signal peak value of any one of the sleeper signal and the track bed signal using the Ricker wavelet template method or the time-domain waveform method; or, to extract the energy peak value in the time-frequency domain of any one signal using the continuous wavelet transform method as the signal peak value of any one signal.

[0163] In one embodiment, the distance is the distance between the top of the sleeper and the top of the track bed. The distance determination module 1630 is specifically used to obtain a preset distance threshold. The preset distance threshold is related to the thickness of the sleeper. Based on whether the distance meets the preset distance threshold, it is determined whether the sleeper is in an empty state during the empty-sleeping detection.

[0164] In one embodiment, the apparatus further includes: a post-processing module, configured to perform zero-bias removal processing on the railway detection signal to obtain a zero-bias removal railway detection signal; perform zero-point adjustment processing on the zero-bias removal railway detection signal to obtain a zero-point adjusted railway detection signal; and perform phase-shift migration processing on the zero-point adjusted railway detection signal in the frequency-wavenumber domain to obtain a migrated railway detection signal.

[0165] Each module in the aforementioned sleeper empty-sleeve status detection device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of the electronic device in hardware form or independent of it, or stored in the memory of the electronic device in software form, so that the processor can call and execute the corresponding operations of each module.

[0166] In one exemplary embodiment, an electronic device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 17 As shown, the electronic device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a method for detecting the empty status of railway sleepers. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the electronic device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the electronic device, or external keyboards, touchpads, or mice, etc.

[0167] Those skilled in the art will understand that Figure 17 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the electronic device to which the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0168] In one embodiment, an electronic device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0169] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0170] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0171] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0172] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0173] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0174] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for detecting the unloaded state of railway sleepers, characterized in that, The method includes: Acquire railway detection signals; the railway detection signals include detection signals collected by ground-penetrating radar mounted on the railway vehicle when the railway vehicle is running on the railway track; The time-domain waveform difference information between the sleeper signal and the ballast signal at the detection location is determined based on the railway detection signal; the sleeper signal includes the electromagnetic wave signal reflected by the sleeper; the ballast signal includes the electromagnetic wave signal reflected by the ballast. Based on the time-domain waveform difference information, the distance between the sleeper and the track bed at the detection location is determined; Based on the distance, the empty-sleeper detection result at the detection location is determined.

2. The method according to claim 1, characterized in that, The step of determining the time-domain waveform difference information between the sleeper signal and the ballast signal at the detection location based on the railway detection signal includes: Extract the sleeper signal and the track bed signal at the detection location from the railway detection signal; Based on the track bed signal and the sleeper signal, a first time and a second time are determined respectively; the first time includes the time when the ground penetrating radar receives the electromagnetic wave signal reflected from the top of the track bed at the detection position; the second time includes the time when the ground penetrating radar receives the electromagnetic wave signal reflected from the top of the sleeper at the detection position. The time-domain waveform difference information is generated based on the difference between the first time and the second time.

3. The method according to claim 2, characterized in that, The step of determining the first time and the second time based on the track bed signal and the sleeper signal respectively includes: Determine the signal peak values ​​of the sleeper signal and the track bed signal respectively; The sampling time corresponding to the peak value of the track bed signal is taken as the first time. The sampling time corresponding to the peak value of the sleeper signal is taken as the second time.

4. The method according to claim 3, characterized in that, The step of determining the signal peak values ​​of the sleeper signal and the track bed signal respectively includes: For any one of the sleeper signal and the track bed signal, the signal peak value of any one signal is searched using the Ricker wavelet template method or the time-domain waveform method; Alternatively, the energy peak value in the time-frequency domain of any signal can be extracted using the continuous wavelet transform method, and used as the signal peak value of any signal.

5. The method according to claim 2, characterized in that, The distance is the distance between the top of the sleeper and the top of the track bed. Determining the empty sleeper inspection result at the inspection location based on the distance includes: Obtain a preset distance threshold; the preset distance threshold is related to the thickness of the sleeper; Based on whether the distance meets the preset distance threshold, it is determined whether the sleeper is in an empty state during the empty hoisting detection.

6. The method according to claim 1, characterized in that, After the step of acquiring the railway detection signal, the method further includes: The railway detection signal is subjected to zero-bias removal processing to obtain the zero-bias removed railway detection signal; The zero-bias-free railway detection signal is subjected to zero-point adjustment processing to obtain the zero-point adjusted railway detection signal; The zero-point adjusted railway detection signal is subjected to phase shift migration processing in the frequency-wavenumber domain to obtain the migrated railway detection signal.

7. A device for detecting the unloaded state of a railway sleeper, characterized in that, The device includes: The signal acquisition module is used to acquire railway detection signals; the railway detection signals include detection signals collected by ground-penetrating radar mounted on the railway vehicle when the railway vehicle is running on the railway track. The information determination module is used to determine the time-domain waveform difference information between the sleeper signal and the ballast signal at the detection location based on the railway detection signal; the sleeper signal includes the electromagnetic wave signal reflected by the sleeper; the ballast signal includes the electromagnetic wave signal reflected by the ballast. The distance determination module is used to determine the distance between the sleeper and the track bed at the detection location based on the time-domain waveform difference information. The result determination module is used to determine the empty-sleeper detection result at the detection location based on the distance.

8. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images