A data management method for railway infrastructure maintenance

Abstract

The present invention relates to the technical field of data management, and in particular to a data management method for railway infrastructure maintenance. The method obtains at least two different parameter combinations including a probe type, an ultrasonic frequency, and an incident angle during ultrasonic emission, and obtains an ultrasonic A-scan image corresponding to each parameter combination; obtains an initial abnormality rate of a railway track according to a wave peak in each ultrasonic A-scan image and a time interval between adjacent wave peaks; if the initial abnormality rate is greater than a preset abnormality rate threshold, obtains an overall detection interference degree; if the overall detection interference degree is less than a preset interference degree threshold, selects an optimal ultrasonic A-scan image from all ultrasonic A-scan images according to the global reliability of each ultrasonic A-scan image, stores the optimal ultrasonic A-scan image, and can quickly extract key information in the maintenance of the railway track.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and in particular to a data management method for railway infrastructure maintenance. Background Art

[0002] The maintenance of railway infrastructure is the key to ensure the safety, smoothness and long-term operation of railway transportation. Railway infrastructure includes tracks, bridges, tunnels, power equipment and communication facilities. In order to ensure the safety and efficiency of railways, these facilities must be regularly inspected and maintained, and the maintenance data must be backed up and saved. As the most important part of maintenance, railway tracks are currently collected in real time by various sensors and detection instruments equipped by track inspection vehicles while they are traveling along the railway route. However, due to the detection limitations of track inspection vehicles, track inspection vehicles can only detect the geometry and infrastructure of the track, and the detection capabilities of internal defects of the track (such as cracks, pores, fatigue damage, etc.) are very limited, so they cannot provide in-depth analysis and evaluation. Therefore, it is necessary to combine other detection technologies (such as ultrasonic detection technology) to supplement them, so as to be able to fully detect the track status and ensure the safe operation of the railway.

[0003] As a non-destructive testing technology, ultrasonic testing can very accurately detect and locate defects and damage on the surface and inside of rails. In conjunction with track inspection vehicles, it can comprehensively detect the track status. However, when using ultrasonic testing, since the working principle of ultrasonic equipment depends on the transmission and reception of electrical signals, the power system on the railway line (such as high-pressure power transmission lines, power supply equipment, etc.) or the electrical equipment of the railway train itself may generate a strong electromagnetic field, which will cause the received echo signal to be distorted. At the same time, when ultrasonic waves encounter defects or uneven track structures (such as welds, welded joints), they will produce diffraction effects. Both of these situations will cause more echoes and pseudo-peaks to appear in the acquired ultrasonic A-scan images, and the arrival time of the echo signals will shift and the shape will change. If such ultrasonic A-scan images are compressed and saved, misleading analysis results will be generated, resulting in missed detection, misdetection or misdiagnosis of defects in the railway track, thereby affecting the safety of railway operation and maintenance decisions.

[0004] Therefore, how to improve the quality of ultrasonic A-scan images to reduce the error in railway track defect detection has become an urgent problem to be solved. Summary of the invention

[0005] In view of this, an embodiment of the present invention provides a data management method for railway infrastructure maintenance to solve the problem of how to improve the quality of ultrasonic A-scan images to reduce the error of railway track defect detection.

[0006] An embodiment of the present invention provides a data management method for railway infrastructure maintenance, the method comprising the following steps:

[0007] Acquire at least two different parameter combinations including probe type, ultrasonic frequency and incident angle during ultrasonic emission, perform ultrasonic detection on the railway track, and obtain an ultrasonic A-scan image corresponding to each parameter combination, wherein the ultrasonic A-scan image is a curve graph with amplitude on the vertical axis and time on the horizontal axis;

[0008] According to the peaks in each of the ultrasonic A-scan images and the time intervals between adjacent peaks, the initial abnormality degree of the corresponding ultrasonic A-scan image is obtained, and according to the initial abnormality degree of each of the ultrasonic A-scan images, the initial abnormality rate of the railway track is obtained;

[0009] If the initial abnormality rate is greater than a preset abnormality rate threshold, then the overall detection interference degree corresponding to all ultrasonic A-scan images is calculated according to the amplitude and number of peaks in each of the ultrasonic A-scan images; if the overall detection interference degree is less than the preset interference degree threshold, the global reliability of the corresponding ultrasonic A-scan image is obtained according to the difference between the local peak fluctuation and the overall peak fluctuation of each of the ultrasonic A-scan images;

[0010] According to the global reliability of each of the ultrasonic A-scan images, the best ultrasonic A-scan image is selected from all the ultrasonic A-scan images for storage.

[0011] Preferably, the obtaining of the initial abnormality degree of the corresponding ultrasonic A-scan image according to the peaks in each ultrasonic A-scan image and the time intervals between adjacent peaks includes:

[0012] For any ultrasonic A-scan image, the number of peaks in the any ultrasonic A-scan image is counted to obtain the difference between the number of peaks and the standard number of peaks, which is recorded as the number difference; according to the time corresponding to each peak in the any ultrasonic A-scan image, the time interval between two adjacent peaks is obtained to obtain the time interval variance;

[0013] The product between the quantity difference and the time interval variance is obtained, and the product is used as an independent variable of a preset hyperbolic tangent function to obtain the initial abnormality degree of any ultrasonic A-scan image.

[0014] Preferably, obtaining the initial abnormality rate of the railway track according to the initial abnormality degree of each ultrasonic A-scan image includes:

[0015] The initial abnormality degree of each ultrasonic A-scan image is compared with a preset initial abnormality threshold, and the number of ultrasonic A-scan images corresponding to the initial abnormality degree greater than the preset initial abnormality threshold is counted and recorded as the target number. The initial abnormality rate of the railway track is obtained based on the ratio between the target number and the total number of all ultrasonic A-scan images.

[0016] Preferably, the calculating the overall detection interference degree corresponding to all ultrasonic A-scan images according to the amplitude and number of peaks in each of the ultrasonic A-scan images comprises:

[0017] The number of peaks in each of the ultrasonic A-scan images is counted respectively to obtain the variance of all the peak numbers, the sum of the amplitudes of all the peaks in each of the ultrasonic A-scan images is calculated respectively to obtain the variance of the sum of all the amplitudes, and the sum of the variance of the number of all peaks and the variance of the sum of all the amplitudes is used as the independent variable of a preset hyperbolic tangent function to obtain the overall detection interference degree corresponding to all the ultrasonic A-scan images.

[0018] Preferably, the step of obtaining the global reliability of the corresponding ultrasonic A-scan image according to the difference between the local peak fluctuation and the overall peak fluctuation of each ultrasonic A-scan image comprises:

[0019] For any ultrasonic A-scan image, obtaining the local reliability of any ultrasonic A-scan image according to the amplitude of the first peak appearing in the any ultrasonic A-scan image;

[0020] Obtaining the overall deviation degree of any ultrasonic A-scan image according to the difference between each wave peak in any ultrasonic A-scan image and a preset baseline;

[0021] The local reliability is used as the independent variable of a preset hyperbolic tangent function to obtain a first mapping value, the negative of the overall deviation is used as the independent variable of a preset exponential function to obtain a second mapping value, and the first mapping value and the second mapping value are weightedly summed to obtain the global reliability of any ultrasonic A-scan image.

[0022] Preferably, obtaining the local reliability of any ultrasonic A-scan image according to the amplitude of the first peak appearing in any ultrasonic A-scan image includes:

[0023] According to the order of appearance of the peaks in any one of the ultrasonic A-scan images, the first peak and the second peak are obtained, the time interval between the time corresponding to the first peak and the time corresponding to the second peak is calculated, the ratio between the time interval and the amplitude of the second peak is calculated, and the amplitude of the first peak and the ratio are weightedly summed to obtain the local reliability of any one of the ultrasonic A-scan images.

[0024] Preferably, obtaining the overall deviation degree of any ultrasonic A-scan image according to the difference between each wave peak in any ultrasonic A-scan image and a preset baseline comprises:

[0025] Constructing a preset baseline in any one of the ultrasonic A-scan images, obtaining the number of intersections between the preset baseline and any one of the ultrasonic A-scan images, calculating the sum of the amplitudes of all the peaks in any one of the ultrasonic A-scan images, obtaining a ratio between the sum of the amplitudes and the number of intersections, and obtaining a product value between the ratio and the number of peaks in any one of the ultrasonic A-scan images;

[0026] Obtain the reciprocal of the time interval between every two adjacent wave peaks in any one of the ultrasonic A-scan images, obtain the cumulative sum of the reciprocals of all time intervals, use the cumulative sum as the independent variable of a preset logarithmic function to obtain the corresponding logarithmic value, add the product value and the logarithmic value to obtain the overall deviation degree of any one of the ultrasonic A-scan images.

[0027] Preferably, after obtaining the initial abnormality rate of the railway track, the method further includes:

[0028] If the initial abnormality rate is less than or equal to a preset abnormality rate threshold, an ultrasonic A-scan image corresponding to the smallest initial abnormality degree is selected from all ultrasonic A-scan images as the best ultrasonic A-scan image of the railway track, and the best ultrasonic A-scan image is stored.

[0029] Preferably, after obtaining the overall detection interference degree, the method further includes:

[0030] If the overall detection interference level is greater than or equal to the preset interference level threshold, a new ultrasonic A-scan image is reacquired under each parameter combination, and the method for acquiring the overall detection interference level is repeated until the overall detection interference level is less than the preset interference level threshold, and the best ultrasonic A-scan image is selected from all new ultrasonic A-scan images for storage.

[0031] Preferably, the step of selecting the best ultrasonic A-scan image from among all the ultrasonic A-scan images for storage according to the global reliability of each ultrasonic A-scan image comprises:

[0032] An ultrasonic A-scan image corresponding to the maximum global reliability is acquired as an optimal ultrasonic A-scan image, and the optimal ultrasonic A-scan image is stored.

[0033] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0034] The present invention obtains at least two different parameter combinations including probe type, ultrasonic frequency and incident angle during ultrasonic emission, performs ultrasonic detection on railway tracks, and obtains ultrasonic A-scan images corresponding to each parameter combination, wherein the ultrasonic A-scan image is a curve graph with amplitude on the vertical axis and time on the horizontal axis; obtains the initial abnormality degree of the corresponding ultrasonic A-scan image according to the peaks in each of the ultrasonic A-scan images and the time interval between adjacent peaks, and obtains the initial abnormality rate of the railway track according to the initial abnormality degree of each of the ultrasonic A-scan images; if the initial abnormality rate is greater than a preset abnormality rate threshold, calculates the overall detection interference degree corresponding to all ultrasonic A-scan images according to the amplitude and number of peaks in each of the ultrasonic A-scan images; if the overall detection interference degree is less than the preset interference degree threshold, obtains the global reliability of the corresponding ultrasonic A-scan image according to the difference between the local peak fluctuation and the overall peak fluctuation of each of the ultrasonic A-scan images; and selects the best ultrasonic A-scan image from all the ultrasonic A-scan images for storage according to the global reliability of each of the ultrasonic A-scan images. Among them, by analyzing the degree of influence of electromagnetic interference on the ultrasonic A-scan image under various parameter combinations, the overall detection interference degree of the ultrasonic detection of the railway track is obtained. When it is determined that the interference degree has little influence on the detection result or is not affected by electromagnetic interference according to the overall detection interference degree, the best ultrasonic A-scan image is selected as the final ultrasonic detection result. This image can best reflect the current detection status of the railway track. The image is compressed and saved, which improves the efficiency of data storage and management. In the subsequent maintenance of the railway track, key information can be quickly extracted to help the maintenance team quickly identify the key health status of the track and support a more efficient decision-making process. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, 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.

[0036] Figure 1 It is a method flow chart of a data management method for railway infrastructure maintenance provided in Embodiment 1 of the present invention. DETAILED DESCRIPTION

[0037] Embodiments of the present disclosure are described in detail below, and examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present disclosure, but should not be understood as limiting the present disclosure.

[0038] It should be noted that the terms "first", "second", etc. in the specification of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure.

[0039] In order to illustrate the technical solution of the present invention, specific embodiments are provided below for illustration.

[0040] See also Figure 1 , is a method flow chart of a data management method for railway infrastructure maintenance provided by Embodiment 1 of the present invention, such as Figure 1 As shown, the method may include:

[0041] Step S101, obtaining at least two different parameter combinations including probe type, ultrasonic frequency and incident angle during ultrasonic emission, performing ultrasonic detection on the railway track, and obtaining an ultrasonic A-scan image under each parameter combination, wherein the ultrasonic A-scan image is a curve graph with amplitude on the vertical axis and time on the horizontal axis.

[0042] First, A-scan processes the received ultrasonic signal into a waveform image, and judges whether there are abnormalities and defects in the object under test, as well as the location and size of the defects based on the shape of the waveform. This scanning method is mainly used for industrial detection. Secondly, all ultrasonic instruments generally record two basic parameters of the echo: the size of the echo (i.e., the amplitude) and the position of the echo relative to the zero point on the time axis (the transmission time in the pulse). Therefore, in the embodiment of the present invention, ultrasonic flaw detectors such as USM Go+, Sonatest iPlex, etc. are used, and the A-scan method is adopted to perform ultrasonic detection on the railway track, so as to detect the defects of the railway track based on the A-scan data.

[0043] Considering that the state, material and environmental conditions of the railway track (such as temperature, humidity, corrosion, etc.) will affect the propagation and reflection of the ultrasonic signal, the embodiment of the present invention adopts a combination of multiple probes, ultrasonic frequencies and incident angles during ultrasonic emission, which are suitable for different railway track conditions and can extract the most accurate and meaningful information from the complex and changeable detection environment. Secondly, when ultrasonic waves encounter defects or the influence of track structure, diffraction effects will inevitably occur, which may cause confusion in the echo signal and affect the location and evaluation of defects. The use of multiple combinations can try to reduce the interference of diffraction effects and obtain more real and effective echo signals. Among them, the parameter combination corresponding to the probe type, ultrasonic frequency and incident angle during ultrasonic emission can be as shown in Table 1:

[0044] Table 1. Parameter combinations

[0045]

[0046] The ultrasonic probe is gently placed on the surface of the railway track to ensure that the probe is in good contact with the surface of the railway track. The railway track is ultrasonically inspected based on the above parameter combination to obtain T ultrasonic A-scan images of the railway track, wherein the ultrasonic A-scan image is a curve graph with amplitude on the vertical axis and time on the horizontal axis.

[0047] It should be noted that the value of T is the same as the number of parameter combinations, and there is no restriction here, which can be set according to the implementation scenario. Before ultrasonic testing of railway tracks, clean the surface of the railway tracks to remove dirt, oil stains, etc. to ensure that the signal can be transmitted smoothly; calibrate the ultrasonic flaw detector to ensure that the instrument is in normal working condition, and apply appropriate coupling agent between the probe and the track surface to ensure effective transmission of ultrasonic signals.

[0048] Step S102, obtaining the initial abnormality degree of the corresponding ultrasonic A-scan image according to the peaks in each ultrasonic A-scan image and the time intervals between adjacent peaks, and obtaining the initial abnormality rate of the railway track according to the initial abnormality degree of each ultrasonic A-scan image.

[0049] During the ultrasonic detection of railway tracks, in addition to the internal defects of the railway tracks themselves, there are also unavoidable diffraction effects and possible electromagnetic interference, which will cause more echoes or pseudo peaks to appear in the ultrasonic A-scan image and affect the shape and time position of the echo. Therefore, the embodiment of the present invention first performs an initial abnormality judgment on T ultrasonic A-scan images to determine whether there are defects in the railway track or whether the ultrasonic detection is interfered.

[0050] Ideally, the ultrasonic A-scan image corresponding to a defect-free, structurally intact and uniform railway track should have three peaks, two reflected waves, the peaks of the direct wave and the positions of the subsequent reflected waves should be relatively consistent in time, and the time intervals between the reflected waves should be relatively uniform. Although such an ideal state cannot exist in actual detection, the difference between the ultrasonic A-scan image without defects and interference and these features in the ideal state will not be very large. When there are defects or interference, these features in the ideal state will be seriously destroyed, and more reflected waves or interference peaks will appear, the time intervals between the reflected waves will be irregular, and the appearance will be more chaotic. Therefore, first, according to the peaks in each ultrasonic A-scan image and the time intervals between adjacent peaks, the initial abnormality degree of the corresponding ultrasonic A-scan image is obtained, and then, according to the initial abnormality degree of each ultrasonic A-scan image, the initial abnormality rate of the railway track is obtained.

[0051] Wherein, according to the peaks in each ultrasonic A-scan image and the time intervals between adjacent peaks, obtaining the initial abnormality degree of the corresponding ultrasonic A-scan image includes:

[0052] For any ultrasonic A-scan image, the number of peaks in the any ultrasonic A-scan image is counted to obtain the difference between the number of peaks and the standard number of peaks, which is recorded as the number difference; according to the time corresponding to each peak in the any ultrasonic A-scan image, the time interval between two adjacent peaks is obtained to obtain the time interval variance;

[0053] The product between the quantity difference and the time interval variance is obtained, and the product is used as an independent variable of a preset hyperbolic tangent function to obtain the initial abnormality degree of any ultrasonic A-scan image.

[0054] In one embodiment, taking the ith ultrasonic A-scan image as an example, the expression for calculating the initial abnormality degree of the ith ultrasonic A-scan image is:

[0055]

[0056] in, represents the initial abnormality of the ith ultrasonic A-scan image, tanh() represents the hyperbolic tangent function, represents the number of peaks in the i-th ultrasonic A-scan image, 3 represents the standard number of peaks, that is, the number of peaks under ideal conditions, It represents the variance of the time interval between every two adjacent wave peaks in the i-th ultrasound A-scan image.

[0057] It should be noted that tanh() is used to limit the result to the range of (0, 1). It is used to characterize the difference between the number of peaks in the ith ultrasonic A-scan image and the number of peaks in the ideal state. The larger the value, the more pseudo peaks caused by reflected waves or electromagnetic interference caused by defects appear in the i-th ultrasonic A-scan image, and the greater the initial abnormality of the i-th ultrasonic A-scan image; in the state without defects and interference, the time intervals of each reflected wave (peak) are relatively uniform, reflecting the consistency of the track structure, while defects or interference will destroy such peak distribution. Therefore, using Characterizes the uniformity of the time interval, The larger the value of , the more it indicates that there is a defect or electromagnetic interference in the i-th ultrasonic A-scan image, and the greater the initial abnormality of the i-th ultrasonic A-scan image.

[0058] Similarly, according to the method for obtaining the initial abnormality level of the i-th ultrasonic A-scan image, the initial abnormality level of each ultrasonic A-scan image can be obtained.

[0059] Wherein, according to the initial abnormality degree of each ultrasonic A-scan image, the initial abnormality rate of the railway track is obtained, including:

[0060] The initial abnormality degree of each ultrasonic A-scan image is compared with a preset initial abnormality threshold, and the number of ultrasonic A-scan images corresponding to the initial abnormality degree greater than the preset initial abnormality threshold is counted and recorded as the target number. The initial abnormality rate of the railway track is obtained based on the ratio between the target number and the total number of all ultrasonic A-scan images.

[0061] In one embodiment, the initial abnormality threshold is set to 0.6. When the initial abnormality degree of any ultrasonic A-scan image is greater than 0.6, it means that the ultrasonic A-scan image is less consistent with the image collected under the ideal state, and is more likely to be caused by defects in the railway track or electromagnetic interference. Therefore, the initial abnormality degree of each ultrasonic A-scan image is compared with the initial abnormality threshold, and the number of ultrasonic A-scan images corresponding to the initial abnormality degree greater than the preset initial abnormality threshold is counted, and recorded as F. Then, according to the number of ultrasonic A-scan images corresponding to the initial abnormality degree greater than the preset initial abnormality threshold, the abnormal situation of the railway track reflected by these T ultrasonic A-scan images is determined, and the calculation expression of the initial abnormality rate of the railway track is:

[0062]

[0063] Wherein, G represents the initial abnormality rate of the railway track, F represents the number of ultrasonic A-scan images corresponding to the initial abnormality degree being greater than a preset initial abnormality threshold, and T represents the total number of ultrasonic A-scan images.

[0064] It should be noted that the higher the initial abnormality rate, the more likely it is that the railway track will have defects or be subject to electromagnetic interference during detection. Therefore, the abnormality rate threshold is set to 0.5. If the initial abnormality rate of the railway track is greater than 0.5, it is considered that the railway track has defects or electromagnetic interference, and it is necessary to further analyze the degree of interference of the ultrasonic A-scan image to obtain the best ultrasonic A-scan image; conversely, if the initial abnormality rate of the railway track is less than or equal to 0.5, the ultrasonic A-scan image corresponding to the smallest initial abnormality degree is selected from all ultrasonic A-scan images as the best ultrasonic A-scan image of the railway track, and the best ultrasonic A-scan image is stored.

[0065] Step S103: If the initial abnormality rate is greater than the preset abnormality rate threshold, the overall detection interference degree corresponding to all ultrasonic A-scan images is calculated according to the amplitude and number of peaks in each ultrasonic A-scan image. If the overall detection interference degree is less than the preset interference degree threshold, the global reliability of the corresponding ultrasonic A-scan image is obtained according to the difference between the local peak fluctuation and the overall peak fluctuation of each ultrasonic A-scan image.

[0066] After determining that the initial abnormality rate of the railway track is greater than 0.5, it can only be said that the railway track may have defects or be subject to electromagnetic interference. Therefore, the degree of electromagnetic interference during this detection process is further analyzed. In the case of strong electromagnetic interference, the intensity of the interference source will fluctuate at different times, which leads to large differences in T test results and confusion, specifically manifested as an uncertain number of pseudo peaks and relatively random amplitudes. Therefore, the overall detection interference degree corresponding to all ultrasonic A-scan images can be calculated based on the amplitude and number of peaks in each ultrasonic A-scan image. The specific method is:

[0067] The number of peaks in each of the ultrasonic A-scan images is counted respectively to obtain the variance of all the peak numbers, the sum of the amplitudes of all the peaks in each of the ultrasonic A-scan images is calculated respectively to obtain the variance of the sum of all the amplitudes, and the sum of the variance of the number of all peaks and the variance of the sum of all the amplitudes is used as the independent variable of a preset hyperbolic tangent function to obtain the overall detection interference degree corresponding to all the ultrasonic A-scan images.

[0068] In one embodiment, the calculation expression of the overall detection interference degree is:

[0069]

[0070] Where H represents the overall detection interference level, tanh() represents the hyperbolic tangent function, represents the number of peaks in the i-th ultrasonic A-scan image, represents the variance of the number of all peaks, represents the sum of the amplitudes of all peaks in the i-th ultrasonic A-scan image, It represents the variance of the sum of the amplitudes of all ultrasound A-scan images.

[0071] It should be noted that the more peaks there are in any ultrasonic A-scan image, the greater the electromagnetic interference. The greater the variance of the peaks between all ultrasonic A-scan images, the different degrees of electromagnetic interference between ultrasonic A-scan images. The greater the variance, the more serious the electromagnetic interference, and the greater the corresponding overall detection interference. Characterizes the sum of the amplitudes of all peaks in the i-th ultrasonic A-scan image, using the variance To characterize the peak amplitude difference between all ultrasound A-scan images, The larger the value is, the more serious the electromagnetic interference is, and the greater the corresponding overall detection interference level is.

[0072] The greater the overall detection interference level, the more unstable the signals are during multiple groups of ultrasonic detection, resulting in greater differences in multiple groups of detection results, and the more likely they are to be affected by electromagnetic interference. Therefore, the interference level threshold is set to 0.6. When the overall detection interference level is in the range of [0.6, 1], it is determined that the railway track detection is subject to strong electromagnetic interference, and shielding technology is required for re-detection. Specifically: if the overall detection interference level is greater than or equal to the preset interference level threshold, a new ultrasonic A-scan image is re-acquired under each of the parameter combinations, and the method for acquiring the overall detection interference level is repeated until the overall detection interference level is less than the preset interference level threshold, and the best ultrasonic A-scan image is selected from all new ultrasonic A-scan images for storage.

[0073] If the overall detection interference level is within the range of (0, 0.6), it means that the electromagnetic interference during railway track detection is low and negligible, and no re-detection is required. Then, by analyzing the global reliability of each ultrasonic A-scan image, a best ultrasonic A-scan image is selected from all ultrasonic A-scan images for storage. The global reliability of each ultrasonic A-scan image is obtained as follows:

[0074] (1) For any ultrasonic A-scan image, the local reliability of the ultrasonic A-scan image is obtained according to the amplitude of the first peak appearing in the ultrasonic A-scan image.

[0075] Considering that the location, size and shape of the track defect are unknown, and the impact of electromagnetic interference and diffraction is relatively complex, the impact of interference on a global scale may be covered by the intensity of other signals, resulting in insufficient reliability of the analysis results. The first two peaks in the ultrasonic A-scan image, namely the direct wave and the outer surface reflection wave, are the strongest parts of the signal and are most affected by electromagnetic interference. The changes in interference on these peaks are more obvious and easier to detect. Therefore, based on the first peak, the local interference degree of any ultrasonic A-scan image is constructed, where the method for obtaining the local interference degree is:

[0076] According to the order of appearance of the peaks in any one of the ultrasonic A-scan images, the first peak and the second peak are obtained, the time interval between the time corresponding to the first peak and the time corresponding to the second peak is calculated, the ratio between the time interval and the amplitude of the second peak is calculated, and the amplitude of the first peak and the ratio are weightedly summed to obtain the local reliability of any one of the ultrasonic A-scan images.

[0077] In one embodiment, taking the ith ultrasonic A-scan image as an example, the calculation expression of the local reliability of the ith ultrasonic A-scan image is:

[0078]

[0079] in, represents the local reliability of the ith ultrasonic A-scan image, represents the first weight, represents the second weight, represents the amplitude of the first peak, represents the amplitude of the second peak, Indicates the time corresponding to the second peak, Indicates the time corresponding to the first peak.

[0080] It should be noted that due to the influence of electromagnetic interference, the time position of the peak will be delayed and the amplitude of the peak will decrease. The larger the value is, the more serious the interference is, and the corresponding i-th ultrasonic A-scan image cannot fully reflect the real information of the railway track, and the corresponding local reliability of the i-th ultrasonic A-scan image is smaller; the smaller the amplitude of the first peak and the second peak is, the stronger the electromagnetic interference is, the more unreal the corresponding railway track information is, and the corresponding local reliability of the i-th ultrasonic A-scan image is greater.

[0081] Since electromagnetic interference or diffraction usually has very little effect on the direct wave (i.e. the first peak), the weight setting should ensure that The principle is preferably that in the embodiment of the present invention, .

[0082] (2) Obtaining the overall deviation degree of any one of the ultrasonic A-scan images based on the difference between each wave peak in any one of the ultrasonic A-scan images and a preset baseline.

[0083] Specifically, a preset baseline is constructed in any one of the ultrasonic A-scan images, the number of intersections between the preset baseline and the any one of the ultrasonic A-scan images is obtained, the sum of the amplitudes of all the peaks in the any one of the ultrasonic A-scan images is calculated, the ratio between the sum of the amplitudes and the number of intersections is obtained, and the product value between the ratio and the number of peaks in the any one of the ultrasonic A-scan images is obtained;

[0084] Obtain the reciprocal of the time interval between every two adjacent wave peaks in any one of the ultrasonic A-scan images, obtain the cumulative sum of the reciprocals of all time intervals, use the cumulative sum as the independent variable of a preset logarithmic function to obtain the corresponding logarithmic value, add the product value and the logarithmic value to obtain the overall deviation degree of any one of the ultrasonic A-scan images.

[0085] In one embodiment, considering that the ultrasonic A-scan image is affected by electromagnetic interference or diffraction effects, signal instability and weak echoes caused by pseudo echoes will appear as a whole, and the baseline (when no ultrasonic signal is returned, the receiver will draw a horizontal line in the ultrasonic A-scan image, called the baseline) will fluctuate over a large range. Therefore, firstly, an ideal baseline value s is obtained (modern ultrasonic equipment provides a "baseline adjustment" function, which can be calibrated during initialization to obtain an initial baseline), and then a straight line Y=s parallel to the X-axis is added to the T ultrasonic A-scan images.

[0086] The number of intersections between the preset baseline and the i-th ultrasonic A-scan image is obtained from the i-th ultrasonic A-scan image. At the same time, the time interval between every two adjacent peaks in the i-th ultrasonic A-scan image is counted. Then, according to the number of intersections and the time interval, the overall deviation degree of the i-th ultrasonic A-scan image is calculated:

[0087]

[0088] in, represents the overall deviation degree of the i-th ultrasonic A-scan image, represents the number of intersections between the preset baseline and the i-th ultrasound A-scan image, represents the amplitude of the jth peak in the i-th ultrasonic A-scan image, represents the number of peaks in the i-th ultrasonic A-scan image, represents the time corresponding to the jth peak in the i-th ultrasonic A-scan image, Indicates the time corresponding to the j-1th peak in the i-th ultrasonic A-scan image.

[0089] It should be noted that the reflected wave peak caused by the defect is usually larger than the peak caused by electromagnetic interference or diffraction, and it will cause a large fluctuation in the baseline, which will lead to The larger the value, the greater the overall deviation of the corresponding i-th ultrasonic A-scan image from the ideal state; The larger the value, the lower the degree of electromagnetic interference, and the smaller the overall difference of the corresponding i-th ultrasonic A-scan image; when electromagnetic interference occurs, pseudo peaks will be introduced, and the time interval between corresponding adjacent peaks will become smaller, so The larger the value is, the greater the degree of electromagnetic interference is, and the greater the overall difference of the corresponding i-th ultrasonic A-scan image is.

[0090] (3) Using the local reliability as an independent variable of a preset hyperbolic tangent function to obtain a first mapping value, using the negative of the overall deviation as an independent variable of a preset exponential function to obtain a second mapping value, performing weighted summation on the first mapping value and the second mapping value to obtain a global reliability of any one of the ultrasonic A-scan images.

[0091] In one embodiment, the calculation expression of the global reliability of the i-th ultrasonic A-scan image is:

[0092]

[0093] in, represents the global reliability of the i-th ultrasonic A-scan image, represents the weight of local reliability, tanh() represents the hyperbolic tangent function, represents the local reliability of the ith ultrasonic A-scan image, Indicates the weight of the overall deviation, exp() represents an exponential function with a natural constant as the base, Represents the overall deviation degree of the i-th ultrasonic A-scan image.

[0094] It should be noted that setting , no restriction is made here. The closer the value is to 1, the clearer the i-th ultrasonic A-scan image is, the less redundant peaks there are, and the echo has obvious peaks and positions, which means that it can better represent the actual condition of the railway track.

[0095] Similarly, the global reliability of each ultrasonic A-scan image is obtained.

[0096] Step S104 , selecting the best ultrasonic A-scan image from all ultrasonic A-scan images for storage according to the global reliability of each ultrasonic A-scan image.

[0097] The greater the global reliability, the more the corresponding ultrasonic A-scan image can reflect the real internal situation of the railway track. Therefore, according to the global reliability of each ultrasonic A-scan image, the ultrasonic A-scan image corresponding to the maximum global reliability is obtained as the best ultrasonic A-scan image, and the best ultrasonic A-scan image is stored. The best ultrasonic A-scan image at this time can best show the detection situation of the current railway track. The best ultrasonic A-scan image is compressed and saved, which improves the efficiency of data storage and management. In the subsequent maintenance of the railway track, key information can be quickly extracted to help the maintenance team quickly identify the key health status of the track and support a more efficient decision-making process.

[0098] In summary, the embodiment of the present invention obtains at least two different parameter combinations including probe type, ultrasonic frequency and incident angle during ultrasonic emission, performs ultrasonic detection on the railway track, and obtains an ultrasonic A-scan image corresponding to each parameter combination, wherein the ultrasonic A-scan image is a curve graph with amplitude on the vertical axis and time on the horizontal axis; the initial abnormality degree of the corresponding ultrasonic A-scan image is obtained according to the peaks in each of the ultrasonic A-scan images and the time interval between adjacent peaks, and the initial abnormality rate of the railway track is obtained according to the initial abnormality degree of each of the ultrasonic A-scan images; if the initial abnormality rate is greater than a preset abnormality rate threshold, the overall detection interference degree corresponding to all ultrasonic A-scan images is calculated according to the amplitude and number of peaks in each of the ultrasonic A-scan images; if the overall detection interference degree is less than the preset interference degree threshold, the global reliability of the corresponding ultrasonic A-scan image is obtained according to the difference between the local peak fluctuation and the overall peak fluctuation of each of the ultrasonic A-scan images; according to the global reliability of each of the ultrasonic A-scan images, the best ultrasonic A-scan image is selected from all the ultrasonic A-scan images for storage. Among them, by analyzing the degree of influence of electromagnetic interference on the ultrasonic A-scan image under various parameter combinations, the overall detection interference degree of the ultrasonic detection of the railway track is obtained. When it is determined that the interference degree has little influence on the detection result or is not affected by electromagnetic interference according to the overall detection interference degree, the best ultrasonic A-scan image is selected as the final ultrasonic detection result. This image can best reflect the current detection status of the railway track. The image is compressed and saved, which improves the efficiency of data storage and management. In the subsequent maintenance of the railway track, key information can be quickly extracted to help the maintenance team quickly identify the key health status of the track and support a more efficient decision-making process.

[0099] 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 may be replaced by equivalents. Such 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, and should be included in the protection scope of the present invention.

Claims

1. A data management method for railway infrastructure maintenance, characterized in that: The method comprises: Acquire at least two different parameter combinations including probe type, ultrasonic frequency and incident angle during ultrasonic emission, perform ultrasonic detection on the railway track, and obtain an ultrasonic A-scan image corresponding to each parameter combination, wherein the ultrasonic A-scan image is a curve graph with amplitude on the vertical axis and time on the horizontal axis; According to the peaks in each of the ultrasonic A-scan images and the time intervals between adjacent peaks, the initial abnormality degree of the corresponding ultrasonic A-scan image is obtained, and according to the initial abnormality degree of each of the ultrasonic A-scan images, the initial abnormality rate of the railway track is obtained; If the initial abnormality rate is greater than a preset abnormality rate threshold, then the overall detection interference degree corresponding to all ultrasonic A-scan images is calculated according to the amplitude and number of peaks in each of the ultrasonic A-scan images; if the overall detection interference degree is less than the preset interference degree threshold, the global reliability of the corresponding ultrasonic A-scan image is obtained according to the difference between the local peak fluctuation and the overall peak fluctuation of each of the ultrasonic A-scan images; According to the global reliability of each of the ultrasonic A-scan images, selecting the best ultrasonic A-scan image from all the ultrasonic A-scan images for storage; The step of obtaining the initial abnormality rate of the railway track according to the initial abnormality degree of each ultrasonic A-scan image comprises: The initial abnormality degree of each ultrasonic A-scan image is compared with a preset initial abnormality threshold, and the number of ultrasonic A-scan images corresponding to the initial abnormality degree greater than the preset initial abnormality threshold is counted, recorded as the target number, and the initial abnormality rate of the railway track is obtained according to the ratio between the target number and the total number of all ultrasonic A-scan images; The step of calculating the overall detection interference degree corresponding to all ultrasonic A-scan images according to the amplitude and number of peaks in each ultrasonic A-scan image comprises: The number of peaks in each of the ultrasonic A-scan images is counted respectively to obtain the variance of all the peak numbers, the sum of the amplitudes of all the peaks in each of the ultrasonic A-scan images is calculated respectively to obtain the variance of the sum of all the amplitudes, and the sum of the variance of all the peak numbers and the variance of the sum of all the amplitudes is used as the independent variable of a preset hyperbolic tangent function to obtain the overall detection interference degree corresponding to all the ultrasonic A-scan images; The step of obtaining the global reliability of the corresponding ultrasonic A-scan image according to the difference between the local peak fluctuation and the overall peak fluctuation of each ultrasonic A-scan image comprises: For any ultrasonic A-scan image, obtaining the local reliability of any ultrasonic A-scan image according to the amplitude of the first peak appearing in the any ultrasonic A-scan image; Obtaining the overall deviation degree of any ultrasonic A-scan image according to the difference between each wave peak in any ultrasonic A-scan image and a preset baseline; The local reliability is used as the independent variable of a preset hyperbolic tangent function to obtain a first mapping value, the negative of the overall deviation is used as the independent variable of a preset exponential function to obtain a second mapping value, and the first mapping value and the second mapping value are weightedly summed to obtain the global reliability of any ultrasonic A-scan image.

2. A data management method for railway infrastructure maintenance according to claim 1, characterized in that: The step of obtaining the initial abnormality degree of the corresponding ultrasonic A-scan image according to the peaks in each ultrasonic A-scan image and the time intervals between adjacent peaks includes: For any ultrasonic A-scan image, the number of peaks in the any ultrasonic A-scan image is counted to obtain the difference between the number of peaks and the standard number of peaks, which is recorded as the number difference; according to the time corresponding to each peak in the any ultrasonic A-scan image, the time interval between two adjacent peaks is obtained to obtain the time interval variance; The product between the quantity difference and the time interval variance is obtained, and the product is used as an independent variable of a preset hyperbolic tangent function to obtain the initial abnormality degree of any ultrasonic A-scan image.

3. A data management method for railway infrastructure maintenance according to claim 1, characterized in that: The obtaining of the local reliability of any ultrasonic A-scan image according to the amplitude of the first peak appearing in any ultrasonic A-scan image comprises: According to the order of appearance of the peaks in any one of the ultrasonic A-scan images, the first peak and the second peak are obtained, the time interval between the time corresponding to the first peak and the time corresponding to the second peak is calculated, the ratio between the time interval and the amplitude of the second peak is calculated, and the amplitude of the first peak and the ratio are weightedly summed to obtain the local reliability of any one of the ultrasonic A-scan images.

4. A data management method for railway infrastructure maintenance according to claim 1, characterized in that: The step of obtaining the overall deviation degree of any ultrasonic A-scan image according to the difference between each wave peak in any ultrasonic A-scan image and a preset baseline comprises: Constructing a preset baseline in any one of the ultrasonic A-scan images, obtaining the number of intersections between the preset baseline and any one of the ultrasonic A-scan images, calculating the sum of the amplitudes of all the peaks in any one of the ultrasonic A-scan images, obtaining a ratio between the sum of the amplitudes and the number of intersections, and obtaining a product value between the ratio and the number of peaks in any one of the ultrasonic A-scan images; Obtain the reciprocal of the time interval between every two adjacent wave peaks in any one of the ultrasonic A-scan images, obtain the cumulative sum of the reciprocals of all time intervals, use the cumulative sum as the independent variable of a preset logarithmic function to obtain the corresponding logarithmic value, add the product value and the logarithmic value to obtain the overall deviation degree of any one of the ultrasonic A-scan images.

5. A data management method for railway infrastructure maintenance according to claim 1, characterized in that: After obtaining the initial abnormality rate of the railway track, the method further includes: If the initial abnormality rate is less than or equal to a preset abnormality rate threshold, an ultrasonic A-scan image corresponding to the smallest initial abnormality degree is selected from all ultrasonic A-scan images as the best ultrasonic A-scan image of the railway track, and the best ultrasonic A-scan image is stored.

6. A data management method for railway infrastructure maintenance according to claim 1, characterized in that: After obtaining the overall detection interference level, the method further includes: If the overall detection interference level is greater than or equal to the preset interference level threshold, a new ultrasonic A-scan image is reacquired under each parameter combination, and the method for acquiring the overall detection interference level is repeated until the overall detection interference level is less than the preset interference level threshold, and the best ultrasonic A-scan image is selected from all new ultrasonic A-scan images for storage.

7. A data management method for railway infrastructure maintenance according to claim 1, characterized in that: The step of selecting the best ultrasonic A-scan image from all the ultrasonic A-scan images for storage according to the global reliability of each ultrasonic A-scan image comprises: An ultrasonic A-scan image corresponding to the maximum global reliability is acquired as an optimal ultrasonic A-scan image, and the optimal ultrasonic A-scan image is stored.

Images