Rail welding seam detection method and device, electronic equipment and storage medium

Abstract

The invention provides a track welding seam detection method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining the vibration acceleration data of a to-be-detected track collected by a vehicle-mounted axle box, screening the vibration acceleration data based on a corrugation filtering algorithm of a track impact index, obtaining the track impact index data, and obtaining the track welding seam detection result of the to-be-detected track. Detecting weld joint disease data in the track impact index data to determine and obtain a weld joint disease position of the track to be detected, extracting weld joint characteristic disease vibration data from the vibration acceleration data corresponding to the weld joint disease position through a characteristic disease separation algorithm of self-adjusting wavelet decomposition, and determining the weld joint characteristic disease vibration data according to the weld joint characteristic disease vibration data. And calculating an energy value of the welding seam characteristic disease vibration data so as to detect and obtain welding seam disease deterioration degree information of the to-be-detected track. Therefore, a set of method for detecting track welding seam diseases by using vibration acceleration data is formed, and the problem of inaccurate track welding seam detection results is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of weld defect detection, and in particular to a rail weld detection method, device, electronic equipment and storage medium. Background Art

[0002] As the annual transportation volume of heavy-haul railways continues to rise and the operating time and mileage increase, problems caused by wheel-rail contact emerge in an endless stream, such as damage to mobile equipment and infrastructure components caused by wheel-rail impact, trouble to residents along the line caused by wheel-rail noise, and increased maintenance and repair costs. The unevenness of rail welding joints is the main prominent disease, which will cause a series of prominent wheel-rail dynamic problems, deteriorate the status of heavy-haul railway lines, and further affect the safety and stability of heavy-haul train operation. In severe cases, it may even cause serious safety accidents such as derailment and overturning of heavy-haul trains.

[0003] At present, the heavy-duty railway comprehensive inspection vehicle is mainly used to inspect the status of heavy-duty railway tracks. Equipment for detecting track geometry and vehicle dynamic response is installed on the inspection train. The vehicle dynamic response includes an axle box acceleration sensor, which can collect the acceleration of the axle box in the lateral and vertical directions in real time. However, the operating speed and axle weight of the inspection train are greatly different from those of the actual operating vehicles, and the line status cannot be accurately detected.

[0004] Therefore, studying a weld defect detection method based on rail impact index to quickly and accurately detect rail welds using vibration acceleration data is of great significance for identifying the distribution of rail weld defects, ensuring the safe use of heavy-load railway tracks, and promoting the sustainable development of railway track transportation. Summary of the invention

[0005] The present invention provides a rail weld detection method, device, electronic equipment and storage medium, which are used to solve the problem of inaccurate rail weld detection results and lay an important foundation for the safe transportation of heavy-load railway tracks.

[0006] In a first aspect, the present invention provides a rail weld detection method, comprising:

[0007] Obtain vibration acceleration data of the track to be tested collected by the vehicle-mounted axle box;

[0008] The vibration acceleration data is screened based on a corrugation filtering algorithm of a rail impact index to obtain rail impact index data;

[0009] Detecting weld defect data in the rail impact index data to determine the weld defect location of the rail to be tested;

[0010] Extracting weld characteristic defect vibration data from the vibration acceleration data corresponding to the weld defect position through a characteristic defect separation algorithm of self-adjusting wavelet decomposition;

[0011] The energy value of the weld characteristic defect vibration data is calculated to determine the weld defect degradation degree information of the track to be tested.

[0012] Optionally, the vibration acceleration data is screened based on a corrugation filtering algorithm of a rail impact index to obtain rail impact index data, including:

[0013] According to a preset calculation formula for moving effective value, the vibration acceleration data is normalized to obtain a preliminary track impact index;

[0014] The preset threshold segment is used as the length unit, and the corrugation segment in which the length of abnormal data in the preliminary orbital impact index is greater than the length unit is eliminated to obtain the orbital impact index data; the abnormal data is the data in the preliminary orbital impact index that is greater than the preset corrugation threshold; the abnormal data length is the data length of the abnormal data that appears continuously when the abnormal data appears in the preliminary orbital impact index.

[0015] Optionally, the vibration acceleration data is normalized according to a preset moving effective value calculation formula to obtain a preliminary track impact index, including:

[0016] According to the preset movement effective value calculation formula, multiple movement effective values ​​in the vibration acceleration data are calculated;

[0017] The preliminary orbital impact index is obtained based on a ratio of a plurality of the moving effective values ​​to an average value of the plurality of the moving effective values.

[0018] Optionally, using a preset threshold value segmentation segment as a length unit, eliminating the corrugation segment whose abnormal data length in the preliminary track impact index is greater than the length unit, to obtain the track impact index data, including:

[0019] Using the preset threshold segment as a length unit, detecting the data length of the abnormal data in the preliminary orbital impact index to determine whether the data length of each abnormal data in the preliminary orbital impact index is greater than the length unit;

[0020] If yes, determine that the abnormal data with a data length greater than the length unit is a corrugation section, remove the corrugation section from the preliminary track impact index, and obtain the track impact index data;

[0021] If not, it is determined that the abnormal data in the preliminary track impact index is a non-corrugation section, and the track impact index data is obtained.

[0022] Optionally, detecting weld defect data in the rail impact index data to determine the weld defect location of the rail to be tested includes:

[0023] According to a preset calculation formula for the unevenness threshold value, preliminary weld defect data in the track impact index data is calculated;

[0024] Selecting data whose corresponding moving effective value is greater than a preset 3σ extraction threshold from the preliminary weld defect data to obtain the weld defect data;

[0025] The track position of the weld defect data on the track to be tested is determined as the weld defect position.

[0026] Optionally, the characteristic defect vibration data of the weld is extracted from the vibration acceleration data corresponding to the weld defect position by a characteristic defect separation algorithm of self-adjusting wavelet decomposition, including:

[0027] Extracting the vibration acceleration data corresponding to the weld defect position and recording it as weld vibration acceleration data;

[0028] According to a preset wavelet basis function, the weld vibration acceleration data is converted into corresponding wavelet coefficients, and the wavelet coefficients greater than a preset wavelet threshold are extracted as data containing only defect characteristic components to obtain effective fault signal data;

[0029] The non-zero wavelet coefficients are extracted from the effective fault signal data to obtain the weld characteristic disease vibration data.

[0030] In a second aspect, the present invention provides a rail weld detection device, comprising:

[0031] Acquisition module, used to obtain the vibration acceleration data of the track to be tested collected by the vehicle axle box

[0032] A screening module, used for screening the vibration acceleration data based on a corrugation filtering algorithm of a rail impact index to obtain rail impact index data;

[0033] A detection module, used to detect weld defect data in the rail impact index data, so as to determine the weld defect position of the rail to be tested;

[0034] An extraction module, used for extracting weld characteristic disease vibration data from the vibration acceleration data corresponding to the weld disease position by using a characteristic disease separation algorithm of self-adjusting wavelet decomposition;

[0035] The calculation module is used to calculate the energy value of the weld characteristic defect vibration data to determine the weld defect degradation degree information of the track to be tested.

[0036] In a third aspect, the present invention provides an electronic device, comprising a processor and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the steps in the method provided in the first aspect are performed.

[0037] In a fourth aspect, the present invention provides a storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, runs the steps of the method provided in the first aspect.

[0038] In a fifth aspect, the present invention provides a computer program product, comprising a computer program / instructions, wherein when the computer program is executed by a processor, the steps in the method provided in the first aspect are performed.

[0039] It can be seen from the above technical solutions that the present invention has the following advantages:

[0040] The present invention provides a rail weld detection method, device, electronic device and storage medium. The method includes: obtaining vibration acceleration data of the track to be tested collected by the on-board axle box, filtering the vibration acceleration data based on the wave grinding filtering algorithm of the track impact index to obtain the track impact index data, detecting the weld defect data in the track impact index data to determine the weld defect position of the track to be tested, extracting the weld characteristic defect vibration data from the vibration acceleration data corresponding to the weld defect position through the characteristic defect separation algorithm of self-adjusting wavelet decomposition, calculating the energy value of the weld characteristic defect vibration data to determine the weld defect deterioration degree information of the track to be tested and obtain the rail weld detection value of the corresponding target layer section. A set of methods that can use vibration acceleration data to detect rail weld defects is formed to solve the problem of inaccurate rail weld detection results and lay an important foundation for the safe transportation of heavy-duty railway tracks. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0042] Figure 1 A flow chart of a first embodiment of a rail weld detection method of the present invention;

[0043] Figure 2 This is a flow chart of a second embodiment of a rail weld detection method of the present invention;

[0044] Figure 3 A schematic diagram of vibration acceleration data and track impact index data of the present invention;

[0045] Figure 4 A schematic diagram of track impact index data of the present invention;

[0046] Figure 5 A schematic diagram of weld defect data of the present invention;

[0047] Figure 6 A schematic diagram of a wavelet coefficient of the present invention;

[0048] Figure 7 The present invention is a structural block diagram of a rail weld detection device embodiment of the present invention. DETAILED DESCRIPTION

[0049] The embodiments of the present invention provide a rail weld detection method, device, electronic device and storage medium, which are used to solve the problem of difficulty in detecting rail welds in abnormally high-pressure sections, and lay an important foundation for the safe transportation of heavy-load railway tracks.

[0050] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] Embodiment 1

[0052] See also Figure 1 , Figure 1 The present invention is a flow chart of a first embodiment of a rail weld detection method, the method comprising:

[0053] Step S101, obtaining vibration acceleration data of the track to be tested collected by the vehicle-mounted axle box;

[0054] The embodiment of the present invention obtains vibration acceleration data on the track to be tested collected by an axle box acceleration sensor on a testing train.

[0055] Step S102, filtering the vibration acceleration data based on the corrugation filtering algorithm of the rail impact index to obtain rail impact index data;

[0056] In an optional embodiment, the vibration acceleration data is screened based on a corrugation filtering algorithm of a rail impact index to obtain rail impact index data, including:

[0057] According to a preset calculation formula for moving effective value, the vibration acceleration data is normalized to obtain a preliminary track impact index;

[0058] The preset threshold segment is used as the length unit, and the corrugation segment in which the length of abnormal data in the preliminary orbital impact index is greater than the length unit is eliminated to obtain the orbital impact index data; the abnormal data is the data in the preliminary orbital impact index that is greater than the preset corrugation threshold; the abnormal data length is the data length of the abnormal data that appears continuously when the abnormal data appears in the preliminary orbital impact index.

[0059] The embodiment of the present invention calculates the moving effective value in the vibration acceleration data according to a preset moving effective value calculation formula, obtains a preliminary orbital impact index based on the ratio of the moving effective value to the average value of the moving effective value, divides the segment into segments with a preset threshold value as a length unit, detects the data length of the abnormal data in the preliminary orbital impact index, and determines whether the data length of each abnormal data in the preliminary orbital impact index is greater than the length unit. If so, it is determined that the abnormal data with a data length greater than the length unit is a corrugation segment, and the corrugation segment is eliminated from the preliminary orbital impact index to obtain orbital impact index data. If not, it indicates that the abnormal data in the preliminary orbital impact index is a non-corrugation segment, and orbital impact index data is obtained.

[0060] Step S103, detecting the weld defect data in the rail impact index data to determine the weld defect position of the rail to be tested;

[0061] The embodiment of the present invention obtains weld defect data in the track impact index data through a difference positioning algorithm based on the 3σ criterion, calculates preliminary weld defect data in the track impact index data according to a preset unevenness defect threshold calculation formula, extracts data whose corresponding moving effective value is greater than a preset 3σ extraction threshold from the preliminary weld defect data, obtains weld defect data, and determines the track position of the weld defect data on the track to be tested as the weld defect position.

[0062] Step S104, extracting weld characteristic defect vibration data from the vibration acceleration data corresponding to the weld defect position by using a characteristic defect separation algorithm of self-adjusting wavelet decomposition;

[0063] The embodiment of the present invention extracts the vibration acceleration data corresponding to the weld defect position, recorded as the weld vibration acceleration data, converts the weld vibration acceleration data into the corresponding wavelet coefficients according to a preset wavelet basis function, and extracts data containing only defect characteristic components from the wavelet coefficients based on a preset wavelet threshold to obtain valid fault signal data, and extracts non-zero wavelet coefficients from the valid fault signal data to obtain weld characteristic defect vibration data.

[0064] Step S105, calculating the energy value of the vibration data of the weld characteristic defect to detect and obtain the weld defect degradation degree information of the track to be tested.

[0065] The embodiment of the present invention calculates the energy value of the weld characteristic defect vibration data through a preset energy value calculation formula, and determines the degree of weld defect degradation of the track to be tested by the size of the energy value. The larger the energy value, the more serious the defect degradation.

[0066] A rail weld detection method provided by an embodiment of the present invention includes: obtaining vibration acceleration data of the track to be tested collected by the on-board axle box, filtering the vibration acceleration data based on the wave grinding filtering algorithm of the track impact index to obtain the track impact index data, detecting the weld defect data in the track impact index data to determine the weld defect position of the track to be tested, extracting the weld characteristic defect vibration data from the vibration acceleration data corresponding to the weld defect position through the characteristic defect separation algorithm of the self-adjusting wavelet decomposition, calculating the energy value of the weld characteristic defect vibration data to determine the weld defect degradation degree information of the track to be tested and obtain the rail weld detection value of the corresponding target layer section. A set of methods that can use vibration acceleration data to detect rail weld defects is formed to solve the problem of inaccurate rail weld detection results and lay an important foundation for the safe transportation of heavy-duty railway tracks.

[0067] Embodiment 2

[0068] See also Figure 2 , Figure 2 This is a flow chart of a second embodiment of a rail weld detection method of the present invention, the steps comprising:

[0069] S201, obtaining vibration acceleration data of the track to be tested collected by the vehicle-mounted axle box;

[0070] The embodiment of the present invention obtains vibration acceleration data on the track to be tested from an axle box acceleration sensor on a testing train.

[0071] S202, normalizing the vibration acceleration data according to a preset moving effective value calculation formula to obtain a preliminary track impact index;

[0072] In an optional embodiment, the vibration acceleration data is normalized according to a preset moving effective value calculation formula to obtain a preliminary track impact index, including:

[0073] According to the preset movement effective value calculation formula, multiple movement effective values ​​in the vibration acceleration data are calculated;

[0074] The preliminary orbital impact index is obtained based on a ratio of a plurality of the moving effective values ​​to an average value of the plurality of the moving effective values.

[0075] The embodiment of the present invention uses a moving effective value calculation formula to calculate multiple moving effective values, and compares each moving effective value with an average value of the multiple moving effective values ​​to obtain a preliminary orbital impact index.

[0076] The preset calculation formula for the effective moving value is:

[0077]

[0078] Among them, S r is the moving effective value of the vibration acceleration at the measuring point, N is the total length of the track position, K is the preset moving window length, x is the vibration acceleration data, x i is the vibration acceleration data at the i-th track position.

[0079] S203, using a preset threshold value to divide the segment as a length unit, eliminating the corrugation segment in which the length of abnormal data in the preliminary orbital impact index is greater than the length unit, and obtaining the orbital impact index data; the abnormal data is the data in the preliminary orbital impact index that is greater than a preset corrugation threshold; the abnormal data length is the data length of the abnormal data that appears continuously when the abnormal data appears in the preliminary orbital impact index.

[0080] In an optional embodiment, the preset threshold segment is used as a length unit, and the corrugation segment whose abnormal data length in the preliminary track impact index is greater than the length unit is eliminated to obtain the track impact index data, including:

[0081] Using the preset threshold segment as a length unit, detecting the data length of the abnormal data in the preliminary orbital impact index to determine whether the data length of each abnormal data in the preliminary orbital impact index is greater than the length unit;

[0082] If yes, determine that the abnormal data with a data length greater than the length unit is a corrugation section, remove the corrugation section from the preliminary track impact index, and obtain the track impact index data;

[0083] If not, it is determined that the abnormal data in the preliminary track impact index is a non-corrugation section, and the track impact index data is obtained.

[0084] The embodiment of the present invention needs to give priority to eliminating the corrugation disease. Since there is often more than one disease in the track, the corrugation disease often appears near the track welding unevenness disease. The corrugation disease will affect the detection result of the rail welding. Therefore, it is necessary to eliminate the data containing the corrugation disease from the data. The data containing the corrugation disease and the data of the track welding unevenness disease have different characteristics. The data containing the corrugation disease has continuity. The data containing the corrugation disease is generally called a corrugation section. Therefore, the preset threshold segment is used as the length unit, and the abnormal data with a data length greater than the length unit in the preliminary track impact index is determined as the corrugation section. The corrugation section is eliminated from the preliminary track impact index to obtain the track impact index data. If there is no data with a data length greater than the length unit in the preliminary track impact index, the data does not need to be eliminated, and the preliminary track impact index is determined to be the track impact index data.

[0085] For example, take a simulated vibration acceleration data, which contains a section of corrugation disease (10m long wavelength 180mm wave depth 0.1mm) and two sections of welding joint unevenness disease (parameters are wavelength 100mm, wave depth ±0.5mm respectively), and calculate the rail impact index of this section of data, such as Figure 3 As shown, Figure 3 The a curve at the top is the original vibration acceleration data, and the b curve at the bottom is the track impact index data, where the horizontal axis is the track position in meters and the vertical axis is the acceleration in m / s 2 .from Figure 3 It can be seen that the data containing corrugation disease will have a continuous high index phenomenon, and the duration is the same as the length of the corrugation disease, while the weld disease will only cause local impact phenomenon, so the segmentation technology based on continuous exceeding the preset corrugation threshold is used to achieve the filtering of corrugation disease. When corrugation disease occurs in the line, the calculated track impact index will have an interval that continuously exceeds the preset corrugation threshold, and when the welding unevenness disease occurs, there will be data points that locally exceed the preset corrugation threshold, such as Figure 4 As shown, Figure 4 The c interval marked in the figure is the corrugation section, where the horizontal axis is the track position in meters and the vertical axis is the acceleration in m / s 2 At this point, the corrugated section is removed to eliminate the influence of corrugated disease.

[0086] S204, calculating and obtaining preliminary weld defect data in the track impact index data according to a preset calculation formula for the unevenness defect threshold;

[0087] After obtaining the track impact index data, the embodiment of the present invention obtains the weld defect data in the track impact index data through the difference positioning algorithm based on the 3σ criterion, and calculates the preliminary weld defect data in the track impact index data according to the preset unevenness disease threshold calculation formula. The difference positioning algorithm based on the 3σ criterion is a method for finding anomalies based on the 3σ criterion of the track impact index and the median, and it locates the abnormal position in the track impact index through multiple outlier removal processes. First, based on the assumption that the data only contains random errors, the standard deviation is calculated, and an interval is determined. The interval contains random errors with a certain probability. Errors beyond this interval are considered to be gross errors and should be eliminated. According to the 3σ criterion, the probability of the value being distributed in (μ-3σ, μ+3σ) is 0.9974, that is, most of the normal data are in this interval. When the track has uneven welding disease, it will cause local impact, and when calculating the track impact index of this part of the impact, it will cause range abnormal values. These abnormal values ​​can be regarded as gross errors and eliminated using the 3σ criterion.

[0088] The preset calculation formula for the unevenness threshold is as follows:

[0089] T d =x_median+3σ;

[0090] Among them, T d is the roughness threshold, x_median is the median of vibration acceleration, σ is the variance,

[0091] When x i Greater than T d When setting the track impact index TII i =x_median, which means that the weld defect data is eliminated.

[0092] S205, selecting data whose corresponding moving effective value is greater than a preset 3σ extraction threshold from the preliminary weld defect data to obtain the weld defect data;

[0093] In the embodiment of the present invention, since these weld defect data are range-specific abnormalities and cannot be accurately eliminated at one time, the eliminated data are further subjected to the 3σ criterion error elimination process (i.e., repeating the above steps) until there are no gross errors. Before eliminating these weld defect data, these weld defect data are first marked, and the location of the welding unevenness defect is determined according to the marked weld defect data to obtain the weld defect location.

[0094] S206, determining the track position of the weld defect data on the track to be tested as the weld defect position;

[0095] By performing difference processing on the data after multiple processing and the original data, the location of the rail welding unevenness disease can be preliminarily determined. In order to further determine the accuracy of the abnormal location found, the rail impact index value of the abnormal location is checked, and only the abnormal area with a rail impact index value greater than a certain value is retained, so that the location of the welding unevenness disease can be finally located.

[0096] like Figure 5 As shown, box d is the marked weld defect data, that is, the corresponding position of weld defect data d is the weld defect position, where the horizontal axis is the track position in meters and the vertical axis is the acceleration in m / s 2 .

[0097] S207, extracting the vibration acceleration data corresponding to the weld defect position and recording it as weld vibration acceleration data;

[0098] It should be noted that wavelet transform (WT) is a signal processing technology based on multi-scale analysis, which is used to analyze the frequency components and time domain characteristics of signals or data, and has a wide range of applications in signal and image processing. In the process of wavelet transform, different features in the signal can be obtained by selecting different types of wavelet functions. Wavelet transform can decompose the signal into signal components at different scales, and selectively recombine them according to the scale difference between noise and signal. In the field of wavelet signal processing, wavelet denoising algorithm based on soft threshold and hard threshold is one of the most widely used methods.

[0099] The embodiment of the present invention obtains the original vibration acceleration data corresponding to the located weld defect position to obtain the weld vibration acceleration data. The weld vibration acceleration data contains a high content of weld defect data, but there is still interference from track unevenness and other signals. Therefore, self-adjusting wavelet decomposition is proposed to analyze the weld vibration acceleration data, thereby realizing the separation of characteristic components of the defect.

[0100] S208, converting the weld vibration acceleration data into corresponding wavelet coefficients according to a preset wavelet basis function, and extracting the wavelet coefficients greater than a preset wavelet threshold as data containing only defect characteristic components, to obtain effective fault signal data;

[0101] The embodiment of the present invention converts the weld vibration acceleration data into corresponding wavelet coefficients according to the preset wavelet basis function. Since the frequency distribution of the characteristic vibration signal caused by the weld joint is near a central frequency band, the self-adjusting wavelet decomposition uses hard threshold processing to extract the wavelet coefficients near the central frequency band, that is, the wavelet coefficients of the central frequency band are retained and other coefficients are set to zero, thereby forming a target decoupling filter W, and finally the wavelet coefficients are used to reconstruct the defect characteristic signal. The specific algorithm steps are as follows:

[0102] (1) Select a suitable wavelet basis function u (i.e., a preset wavelet basis function), use the corresponding decomposition scale j to decompose the weld vibration acceleration data and perform discrete wavelet transform to obtain the wavelet coefficients at each scale.

[0103] (2) Set the wavelet threshold Th, perform threshold processing on each wavelet coefficient to retain the wavelet coefficients greater than the wavelet threshold Th, and then reconstruct the other wavelet coefficients (that is, set the other wavelet coefficients to zero), so as to obtain a preliminary effective fault signal.

[0104] (3) Modify the wavelet threshold Th and repeat (2) until the preliminary valid fault signal contains only the data of the disease characteristic components, and then the valid fault signal data is obtained.

[0105] S209, extracting non-zero wavelet coefficients from the effective fault signal data to obtain weld characteristic defect vibration data.

[0106] The embodiment of the present invention obtains the position area where the wavelet coefficient is not zero in the effective fault signal data, records the wavelet coefficient of this position area, and can form the decoupling filter W to achieve defect reconstruction and obtain the weld characteristic defect vibration data.

[0107] like Figure 6 As shown, curve f is the vibration acceleration data, curve g is the weld characteristic disease vibration data, and the weld vibration acceleration data is analyzed by using the preset wavelet basis function to obtain the weld characteristic disease vibration data of curve g, where the horizontal axis is the track position in meters and the vertical axis is the acceleration in m / s 2 .

[0108] S2010, calculating the energy value of the weld characteristic defect vibration data to determine the weld defect degradation degree information of the track to be tested.

[0109] The embodiment of the present invention calculates the energy value of the weld characteristic defect vibration data according to a preset energy value calculation formula, and determines the degree of weld defect degradation of the track to be tested according to the size of the energy value, thereby obtaining the weld defect degradation degree information of the track to be tested. The larger the energy value, the more serious the defect degradation, and the smaller the energy value, the milder the defect degradation.

[0110] The preset energy value calculation formula is as follows:

[0111]

[0112] Where E is the energy value.

[0113] A rail weld detection method disclosed in an embodiment of the present invention includes: obtaining vibration acceleration data of the track to be tested collected by the on-board axle box, filtering the vibration acceleration data based on the wave grinding filtering algorithm of the track impact index to obtain the track impact index data, detecting the weld defect data in the track impact index data to determine the weld defect position of the track to be tested, extracting the weld characteristic defect vibration data from the vibration acceleration data corresponding to the weld defect position through the characteristic defect separation algorithm of the self-adjusting wavelet decomposition, calculating the energy value of the weld characteristic defect vibration data to determine the weld defect degradation degree information of the track to be tested. The problem of inaccurate rail weld detection results is solved, laying an important foundation for the safe transportation of heavy-load railway tracks. At the same time, the method is highly operational, and the process of rail weld detection in actual use is intuitive and clear, and can be easily promoted and applied to the detection process of heavy-load railway track defects.

[0114] Embodiment 3

[0115] See also Figure 7 , Figure 7 This is a structural block diagram of an embodiment of a rail weld detection device of the present invention, the device comprises:

[0116] An acquisition module 301 is used to acquire vibration acceleration data of the track to be tested collected by the vehicle-mounted axle box;

[0117] A screening module 302 is used to screen the vibration acceleration data based on a corrugation filtering algorithm of a track impact index to obtain track impact index data;

[0118] A detection module 303 is used to detect weld defect data in the rail impact index data to determine the weld defect position of the rail to be tested;

[0119] An extraction module 304 is used to extract weld characteristic defect vibration data from the vibration acceleration data corresponding to the weld defect position by using a characteristic defect separation algorithm of self-adjusting wavelet decomposition;

[0120] The calculation module 305 is used to calculate the energy value of the weld characteristic defect vibration data to determine the weld defect degradation degree information of the track to be tested.

[0121] In an optional embodiment, the screening module 302 includes:

[0122] A normalization processing submodule is used to perform normalization processing on the vibration acceleration data according to a preset mobile effective value calculation formula to obtain a preliminary track impact index;

[0123] The elimination submodule is used to divide the segments into sections according to a preset threshold value as a length unit, eliminate the corrugation segments in which the length of abnormal data in the preliminary orbital impact index is greater than the length unit, and obtain the orbital impact index data; the abnormal data is the data in the preliminary orbital impact index that is greater than a preset corrugation threshold; the abnormal data length is the data length of the abnormal data that appears continuously when the abnormal data appears in the preliminary orbital impact index.

[0124] In an optional embodiment, the normalization processing submodule includes:

[0125] A calculation unit, configured to calculate a plurality of movement effective values ​​in the vibration acceleration data according to the preset movement effective value calculation formula;

[0126] A ratio unit is used to obtain the preliminary orbital impact index based on the ratio of the multiple movement effective values ​​to the average value of the multiple movement effective values.

[0127] In an optional embodiment, the elimination submodule includes:

[0128] A judging unit, configured to detect the data length of the abnormal data in the preliminary orbital impact index by using the preset threshold segment as a length unit, so as to judge whether the data length of each abnormal data in the preliminary orbital impact index is greater than the length unit; if so, executing a eliminating unit; if not, executing a determining unit;

[0129] A removal unit, used for determining that the abnormal data whose data length is greater than the length unit is a corrugation section, and removing the corrugation section from the preliminary track impact index to obtain the track impact index data;

[0130] The determination unit is used to determine that the abnormal data in the preliminary track impact index is a non-corrugation section, and obtain the track impact index data.

[0131] In an optional embodiment, the detection module 303 includes:

[0132] A calculation submodule, used for calculating the preliminary weld defect data in the track impact index data according to a preset unevenness defect threshold calculation formula;

[0133] A selection module, used to select data whose corresponding moving effective value is greater than a preset 3σ extraction threshold from the preliminary weld defect data to obtain the weld defect data;

[0134] The determination submodule is used to determine the track position of the weld defect data on the track to be tested as the weld defect position.

[0135] In an optional embodiment, the extraction module 304 includes:

[0136] A first extraction submodule is used to extract the vibration acceleration data corresponding to the weld defect position, which is recorded as weld vibration acceleration data;

[0137] The second extraction submodule is used to convert the weld vibration acceleration data into corresponding wavelet coefficients according to a preset wavelet basis function, and extract the wavelet coefficients greater than a preset wavelet threshold as data containing only defect characteristic components to obtain effective fault signal data;

[0138] The third extraction submodule is used to extract non-zero wavelet coefficients from the effective fault signal data to obtain weld characteristic disease vibration data.

[0139] Embodiment 4

[0140] An embodiment of the present invention further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the processor executes the steps of a rail weld detection method of any embodiment.

[0141] Embodiment 5

[0142] An embodiment of the present invention further provides a computer storage medium on which a computer program is stored. When the computer program is executed by the processor, the steps of a rail weld detection method in any embodiment are implemented.

[0143] Embodiment 6

[0144] An embodiment of the present invention further provides a computer program product having a computer program stored thereon, wherein when the computer program is executed by the processor, the steps of a rail weld detection method of any embodiment are implemented.

[0145] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0146] In the several embodiments provided in the present application, it should be understood that the methods, devices, electronic devices and storage media disclosed in the present invention can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0147] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0148] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0149] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a readable storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned readable storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0150] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A rail weld detection method, characterized in that: include: Obtain vibration acceleration data of the track to be tested collected by the vehicle-mounted axle box; The vibration acceleration data is screened based on a corrugation filtering algorithm of a rail impact index to obtain rail impact index data; Detecting weld defect data in the rail impact index data to determine the weld defect location of the rail to be tested; Extracting weld characteristic defect vibration data from the vibration acceleration data corresponding to the weld defect position through a characteristic defect separation algorithm of self-adjusting wavelet decomposition; The energy value of the weld characteristic defect vibration data is calculated to determine the weld defect degradation degree information of the track to be tested.

2. The rail weld detection method according to claim 1, characterized in that: Based on the corrugation filtering algorithm of the rail impact index, the vibration acceleration data is screened to obtain the rail impact index data, including: According to a preset calculation formula for moving effective value, the vibration acceleration data is normalized to obtain a preliminary track impact index; The preset threshold segment is used as the length unit, and the corrugation segment in which the length of abnormal data in the preliminary orbital impact index is greater than the length unit is eliminated to obtain the orbital impact index data; the abnormal data is the data in the preliminary orbital impact index that is greater than the preset corrugation threshold; the abnormal data length is the data length of the abnormal data that appears continuously when the abnormal data appears in the preliminary orbital impact index.

3. The rail weld detection method according to claim 2, characterized in that: According to the preset calculation formula of moving effective value, the vibration acceleration data is normalized to obtain the preliminary track impact index, including: According to the preset movement effective value calculation formula, multiple movement effective values ​​in the vibration acceleration data are calculated; The preliminary orbital impact index is obtained based on a ratio of a plurality of the moving effective values ​​to an average value of the plurality of the moving effective values.

4. The rail weld detection method according to claim 2, characterized in that: The preset threshold segment is used as a length unit, and the corrugation segment whose abnormal data length in the preliminary track impact index is greater than the length unit is eliminated to obtain the track impact index data, including: Using the preset threshold segment as a length unit, detecting the data length of the abnormal data in the preliminary orbital impact index to determine whether the data length of each abnormal data in the preliminary orbital impact index is greater than the length unit; If yes, determine that the abnormal data with a data length greater than the length unit is a corrugation section, remove the corrugation section from the preliminary track impact index, and obtain the track impact index data; If not, it is determined that the abnormal data in the preliminary track impact index is a non-corrugation section, and the track impact index data is obtained.

5. The rail weld detection method according to claim 1, characterized in that: Detecting the weld defect data in the rail impact index data to determine the weld defect position of the rail to be tested includes: According to a preset calculation formula for the unevenness threshold value, preliminary weld defect data in the track impact index data is calculated; Selecting data whose corresponding moving effective value is greater than a preset 3σ extraction threshold from the preliminary weld defect data to obtain the weld defect data; The track position of the weld defect data on the track to be tested is determined as the weld defect position.

6. The rail weld detection method according to claim 1, characterized in that: The characteristic defect vibration data of the weld is extracted from the vibration acceleration data corresponding to the weld defect position by using a characteristic defect separation algorithm of self-adjusting wavelet decomposition, including: Extracting the vibration acceleration data corresponding to the weld defect position and recording it as weld vibration acceleration data; According to a preset wavelet basis function, the weld vibration acceleration data is converted into corresponding wavelet coefficients, and the wavelet coefficients greater than a preset wavelet threshold are extracted as data containing only defect characteristic components to obtain effective fault signal data; A non-zero wavelet coefficient is extracted from the effective fault signal data to obtain weld characteristic disease vibration data.

7. A rail weld detection device, characterized in that: include: Acquisition module, used to obtain the vibration acceleration data of the track to be tested collected by the vehicle axle box A screening module, used for screening the vibration acceleration data based on a corrugation filtering algorithm of a rail impact index to obtain rail impact index data; A detection module, used to detect weld defect data in the rail impact index data, so as to determine the weld defect position of the rail to be tested; An extraction module is used to extract weld characteristic disease vibration data from the vibration acceleration data corresponding to the weld disease position through a characteristic disease separation algorithm of self-adjusting wavelet decomposition; The calculation module is used to calculate the energy value of the weld characteristic defect vibration data to determine the weld defect degradation degree information of the track to be tested.

8. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the method according to any one of claims 1 to 6 is executed.

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

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

Images