Rail corrugation detection method and device, storage medium and computer program

Abstract

The invention discloses a steel rail corrugation detection method and device, a storage medium and a computer program. The method comprises the steps that detection result data corresponding to at least two detection features used for representing steel rail corrugation are obtained; wherein each detection feature corresponds to one piece of detection result data; determining a corrugation area where corrugation occurs based on the detection result data, and calculating the confidence degree of each corrugation level in the corrugation area; the confidence coefficient is used for representing the probability of corrugation grade occurrence; performing data fusion on the confidence coefficients of the same corrugation level determined by the at least two detection features of the same corrugation area to obtain the fusion confidence coefficient of the same corrugation level; and detecting the occurrence of corrugation in the steel rail section based on the fusion confidence of each corrugation level of each corrugation area. According to the invention, data fusion is carried out on the level confidence of different detection features, corrugation is detected based on the fusion confidence, and the accuracy of corrugation detection is improved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of railway tracks, and in particular to a rail corrugation detection method, device, storage medium and computer program. Background Art

[0002] At present, my country's railway transportation is developing in the direction of heavy-duty freight, intelligent information, and automated detection. Ensuring the safety of railway transportation is the core task of railway transportation. Rails are an indispensable infrastructure in railway transportation. As a basic component of the track structure, they are the basis for determining train operations. In recent years, rail corrugation, as a common surface damage of rails, is specifically manifested as a periodic wave-shaped uneven wear along the longitudinal surface of the rails. The presence of corrugation will cause violent vibrations in the wheel-rail parts, seriously affecting the smoothness of train operation. At the same time, it will greatly reduce the service life of vehicles and track structure components, increase maintenance costs, and even cause train derailment, overturning and other accidents in severe cases. Therefore, detecting rail corrugation and timely warning are of great significance to guiding rail grinding decisions and maintaining train operation safety.

[0003] In the prior art, there are technologies that can realize the identification of rail corrugation, for example, the detection of rail corrugation diseases is realized through axle box vibration acceleration, wheel-rail force technology and track acoustic technology, but these rail corrugation identification algorithms can only roughly identify whether there is corrugation in a certain section and cannot achieve accurate identification. In addition, when applied on site, there are technical problems such as difficulty in obtaining original data, low test efficiency and low accuracy of detection structure. Summary of the invention

[0004] Based on the above technical problems, the present disclosure provides a rail corrugation detection method, device, storage medium and computer program.

[0005] In a first aspect, the present disclosure provides a rail corrugation detection method, comprising:

[0006] Acquire at least two detection result data corresponding to the detection features for characterizing rail corrugation; wherein each detection feature corresponds to one detection result data;

[0007] Determine the corrugation area where corrugation occurs based on each of the detection result data, and calculate the confidence of each corrugation level in the corrugation area; the confidence is used to characterize the probability of occurrence of the corrugation level;

[0008] Performing data fusion on the confidences of the same corrugation level determined by at least two detection features of the same corrugation area to obtain fused confidence of the same corrugation level;

[0009] The occurrence of corrugation in the rail section is detected based on the fused confidence of each corrugation level in each corrugation area.

[0010] In some embodiments, the detection features for characterizing rail corrugation include axle box vibration acceleration, wheel-rail force, and track acoustics.

[0011] In some embodiments, the detection result data includes the starting position and the ending position of each corrugation in the rail section and the number of times each corrugation level occurs;

[0012] The step of respectively determining the corrugation areas where corrugation occurs based on the detection result data and calculating the confidence of each corrugation level of the corrugation area comprises:

[0013] Determine the initial corrugation area of ​​each corrugation according to the starting position and the ending position of each corrugation in the detection result data;

[0014] Based on at least two groups of detection result data corresponding to the same detection feature, multiple groups of corrugation initial areas are obtained;

[0015] Merging the multiple groups of initial corrugation areas of the same corrugation mill to obtain a corrugation mill area;

[0016] Counting the number of occurrences of each corrugation level in the corrugation area;

[0017] The confidence level of each corrugation level in each corrugation area is calculated based on the number of occurrences of each corrugation level in each corrugation area.

[0018] In some embodiments, the step of merging the multiple groups of initial corrugation areas of the same corrugation mill to obtain the corrugation area comprises:

[0019] Detecting whether the multiple groups of initial areas of the same corrugation mill overlap;

[0020] When the multiple groups of corrugation initial areas of the same corrugation mill overlap, the sum of the corrugation initial areas is determined as the corrugation area.

[0021] In some embodiments, the step of merging the multiple groups of initial corrugation areas of the same corrugation mill to obtain the corrugation area further comprises:

[0022] Detecting whether the difference between the starting positions of the plurality of groups of initial areas of the same corrugation mill is within a preset range;

[0023] When the difference between the starting positions of a plurality of groups of the corrugation initial regions of the same corrugation mill is within a preset range, each of the corrugation initial regions is determined as the corrugation region.

[0024] In some embodiments, the step of fusing the confidences of the same corrugation level determined by at least two detection features of the same corrugation area to obtain the fused confidence of the same corrugation level comprises:

[0025] When the number of the detection features is two, data fusion is performed on the confidences of the same corrugation level corresponding to the two detection features of the same corrugation area to obtain the fused confidence of the same corrugation level;

[0026] When the data of the detection feature is N, the confidences of the same corrugation level corresponding to two detection features of the same corrugation area are data fused to obtain a fusion result, and the fusion result is data fused with the confidence of the same corrugation level corresponding to another detection feature, until the data of the confidences of the same corrugation level corresponding to N detection features are fused to obtain the fused confidences of the same corrugation level corresponding to all the detection features; N is an integer greater than or equal to 3.

[0027] In some embodiments, the step of fusing the confidences of the same corrugation level determined by at least two detection features of the same corrugation area to obtain the fused confidence of the same corrugation level comprises:

[0028] Based on the evidence theory, data fusion is performed on the confidence of the same corrugation level determined by at least two detection features of the same corrugation area to obtain the fused confidence of the same corrugation level.

[0029] In a second aspect, the present disclosure provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method described in the above aspect when executing the computer program.

[0030] In a third aspect, the present disclosure provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the method described in the above aspects when executed by a processor.

[0031] In a fourth aspect, the present disclosure provides a computer program product, including a computer program, which implements the steps of the method described in the above aspects when executed by a processor.

[0032] The present invention provides a rail corrugation detection method, device, storage medium and computer program. The corrugation areas where corrugation occurs are determined based on the detection result data corresponding to each detection feature, and the confidence of each corrugation level in the corrugation area is calculated. Then, data fusion is performed on the level confidences of different detection features. Compared with detecting corrugation based on the corrugation level confidence obtained based on a single detection feature, the present invention detects corrugation based on the fused confidence, thereby improving the accuracy of corrugation detection and realizing decision-level diagnosis of rail corrugation diseases. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The present disclosure will be described in more detail below based on embodiments and with reference to the accompanying drawings:

[0034] Figure 1 A schematic diagram of a flow chart of a rail corrugation detection method provided in an embodiment of the present disclosure;

[0035] Figure 2 A technical roadmap of a rail corrugation detection method provided in an embodiment of the present disclosure;

[0036] Figure 3 A simplified flow chart of the in-equipment integration of a rail corrugation detection method provided by an embodiment of the present disclosure;

[0037] Figure 4 A simplified flow chart of the integration of devices in a rail corrugation detection method provided in an embodiment of the present disclosure.

[0038] Figure 5 An internal structural diagram of a computer device provided in an embodiment of the present disclosure.

[0039] In the drawings, the same reference numerals are used for the same components, and the drawings are not drawn to scale. DETAILED DESCRIPTION

[0040] In order to enable those skilled in the art to better understand the technical solution of the present disclosure, and to fully understand and implement how the present disclosure applies technical means to solve technical problems and achieve the corresponding technical effects, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only embodiments of a part of the present disclosure, not all of the embodiments. The embodiments of the present disclosure and the various features in the embodiments can be combined with each other without conflict, and the technical solutions formed are all within the scope of protection of the present disclosure. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without making creative work should fall within the scope of protection of the present disclosure.

[0041] It should be noted that the terms "first", "second", etc. in the specification and claims 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. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices.

[0042] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0043] Embodiment 1

[0044] Figure 1 The following is a flow chart of a rail corrugation detection method provided by an embodiment of the present disclosure. Figure 1 As shown, a rail corrugation detection method comprises:

[0045] Step 110, obtaining detection result data corresponding to at least two detection features for characterizing rail corrugation; wherein each detection feature corresponds to a detection result data.

[0046] Corrugation is a common surface damage of rails. The shape of rail corrugation is mainly sinusoidal and has obvious periodic characteristics. This wave-shaped wear phenomenon is the result of the interaction between the wheel and the rail. The detection features used to characterize rail corrugation refer to characteristic parameters that can directly or indirectly reflect the state, location and severity of rail corrugation.

[0047] In one embodiment, the detection features used to characterize rail corrugation include axle box vibration acceleration, wheel-rail force, track acoustics, rail surface morphology, etc., which are not limited herein. Each detection feature corresponds to a detection result data.

[0048] In one embodiment, each detection result data includes the starting position and the ending position of each corrugation in the rail section and the number of occurrences of each corrugation level.

[0049] In one embodiment, the number of detection result data corresponding to each detection feature can be multiple groups. By obtaining multiple groups of detection result data corresponding to the same detection feature, the comprehensive coverage of the detection result data is improved to avoid missed detection or false detection of the detection result data corresponding to the detection feature.

[0050] Step 120, based on each of the detection result data, respectively determine the corrugation areas where the corrugation occurs, and calculate the confidence of each corrugation level in the corrugation area; the confidence is used to characterize the probability of occurrence of the corrugation level.

[0051] The corrugation grades are mainly divided according to the severity of rail surface wear. In one embodiment, the corrugation grades include three grades A, B and C with the severity ranging from high to low; the detection result data includes the starting position and the ending position of each corrugation, and the number of times each corrugation grade occurs in the corrugation.

[0052] In this embodiment, the corrugation areas where corrugation occurs are determined based on each detection result data, the number of occurrences of each corrugation level in the corrugation area is counted, and the confidence of each corrugation level in each corrugation area is calculated based on the number of occurrences of each corrugation level in the corrugation area and the total number of occurrences of all corrugation levels in the corrugation area.

[0053] In one embodiment, the confidence level of each corrugation level is calculated as follows:

[0054]

[0055] Among them, p(y,x) is the confidence of corrugation level x in area range y, m is the number of times corrugation level x appears in area range y, and M is the total number of times all corrugation levels appear in area range y.

[0056] In one embodiment, the number of groups of detection result data corresponding to each detection feature is multiple groups, then multiple corrugation areas are determined based on the multiple groups of detection result data corresponding to the same detection feature, and data fusion is performed on the multiple corrugation areas corresponding to the same detection feature to obtain the fused corrugation area. For example, the detection feature includes a first detection feature and a second detection feature, wherein the first detection feature corresponds to the first detection result data; and the second detection feature corresponds to the second detection result data. The first corrugation area where each corrugation appears is determined for the first group of first detection result data corresponding to the first detection feature, and the second corrugation area where each corrugation appears is determined for the second group of first detection result data corresponding to the first detection feature. The first corrugation area and the second corrugation area of ​​the same corrugation are fused to finally obtain the corrugation area where each corrugation appears determined for the first detection feature. Since the corrugation area determined based on a single detection result data may have the problem of missed detection or inaccurate corrugation area, the corrugation area finally determined in this embodiment is more accurate than the first corrugation area or the second corrugation area. Similarly, based on the multiple groups of second detection result data of the second detection feature or the multiple groups of third detection result data of the third detection feature, the corrugation areas obtained by fusion can be determined respectively.

[0057] Step 130 , data fusion is performed on the confidences of the same corrugation level determined by at least two detection features of the same corrugation area to obtain a fused confidence of the same corrugation level.

[0058] Multi-source data fusion refers to the integration of data from multiple different data sources into a unified data set to absorb the characteristics of different data sources and extract richer and more accurate information from them than a single data source.

[0059] In this embodiment, the confidences of the same corrugation level corresponding to different detection features are fused to obtain fused confidences, thereby improving the accuracy of corrugation detection.

[0060] In this embodiment, by fusing the confidence levels of the same corrugation level of at least two of the detection features in the same corrugation area, the multi-dimensional characteristics of the same corrugation disease and the correlation between different damages can be analyzed from multiple angles, so as to maximize the value of data collected in various fields, provide data support for making scientific and efficient maintenance decisions, and improve the accuracy of corrugation detection.

[0061] Step 140: detecting the occurrence of corrugation in the rail section based on the fused confidence of each corrugation level in each corrugation area.

[0062] In this embodiment, the occurrence of corrugation in the rail section is detected based on the fused confidence of each corrugation level in each corrugation area to determine the area where corrugation in the rail section occurs and the probability of occurrence of each corrugation level in each corrugation area. Compared with detecting corrugation based on the corrugation level confidence obtained based on a single detection feature, the accuracy of detecting corrugation in this embodiment is higher.

[0063] In this embodiment, the corrugation areas where corrugation occurs are determined based on the detection result data corresponding to each detection feature, and the confidence of each corrugation level in the corrugation area is calculated. Then, data fusion is performed on the level confidences of different detection features. Compared with detecting corrugation based on the corrugation level confidence obtained based on a single detection feature, this embodiment detects corrugation based on the fused confidence, thereby improving the accuracy of corrugation detection and achieving decision-level diagnosis of rail corrugation diseases.

[0064] Embodiment 2

[0065] Based on the above embodiment, the detection features for characterizing rail corrugation include axle box vibration acceleration, wheel-rail force and track acoustics.

[0066] Among them, the axle box vibration acceleration is an important dynamic parameter to characterize rail corrugation. When the train passes through the corrugated track, the unevenness of the rail surface will cause the interaction force between the wheel and the rail to change, which in turn causes the vibration of the axle box. The acceleration sensor installed on the axle box can measure the vibration acceleration signal of the axle box in real time. Based on the vibration acceleration signal of the axle box, the detection result data corresponding to the axle box vibration acceleration can be obtained. The detection result data includes the starting position, end position and number of occurrences of each corrugation level in the rail section.

[0067] Wheel-rail force is the result of the interaction between the wheel and the rail, and is one of the important parameters to characterize the rail corrugation state. When rail corrugation occurs, the wheel will be subjected to additional impact force during rolling, resulting in changes in the wheel-rail force. By measuring the magnitude and direction of the wheel-rail force, the severity of rail corrugation can be indirectly evaluated.

[0068] Track acoustic signals refer to the noise generated when the unevenness of the rail surface causes impact and vibration between the wheel and rail when the train passes through the corrugated track. These noise signals contain rich corrugation characteristic information, such as wavelength, wave depth, and distribution of corrugation.

[0069] In one embodiment, the axle box vibration acceleration signal can be obtained by a vibration acceleration sensor installed on the axle box of the train, and the corresponding detection result data can be obtained based on the axle box vibration acceleration signal. The detection result data includes the starting position, end position and number of occurrences of each corrugation level of the rail section.

[0070] In one embodiment, the wheel-rail force signal can be obtained by a force sensor installed on the wheel-rail of the train, and the corresponding detection result data can be obtained based on the wheel-rail force signal. The detection result data includes the starting position, end position and number of occurrences of each corrugation level of the rail section.

[0071] In one embodiment, the track acoustic signal can be obtained by a microphone installed on the wheel rail of the train, and the corresponding detection result data can be obtained based on the track acoustic information. The detection result data includes the starting position, end position and number of occurrences of each corrugation level of the rail section.

[0072] Embodiment 3

[0073] Based on the above embodiment, the detection result data includes the starting position and the ending position of each corrugation in the rail section and the number of times each corrugation level occurs;

[0074] The step of respectively determining the corrugation areas where corrugation occurs based on the detection result data and calculating the confidence of each corrugation level of the corrugation area comprises:

[0075] Determine the initial corrugation area of ​​each corrugation according to the starting position and the ending position of each corrugation in the detection result data;

[0076] Based on at least two groups of detection result data corresponding to the same detection feature, multiple groups of corrugation initial areas are obtained;

[0077] Merging the multiple groups of initial corrugation areas of the same corrugation mill to obtain a corrugation mill area;

[0078] Counting the number of occurrences of each corrugation level in the corrugation area;

[0079] The confidence level of each corrugation level in each corrugation area is calculated based on the number of occurrences of each corrugation level in each corrugation area.

[0080] In one embodiment, there are multiple groups of detection result data corresponding to each detection feature, and multiple groups of corrugation disease tables are determined based on the multiple groups of detection result data corresponding to the same detection feature. Each corrugation disease table includes the initial corrugation area where each corrugation appears and the number of times each corrugation level appears in the initial corrugation area. The initial corrugation area of ​​the same corrugation in the corrugation disease table determined by the detection result data corresponding to the same detection feature is data fused to obtain a fused corrugation area. For example, a first corrugation defect table is determined for a first group of detection result data corresponding to a first detection feature, the first corrugation defect table includes a first corrugation initial area where each corrugation appears; a second corrugation defect table is determined for a second group of detection result data corresponding to the first detection feature, the second corrugation defect table includes a second corrugation initial area where each corrugation appears; the first corrugation initial area and the second corrugation initial area of ​​the same corrugation are fused, and finally the corrugation area where a corrugation appears is determined. Because the corrugation area determined based on a single detection result data may be missed or the corrugation area is inaccurate, the finally determined corrugation area is more accurate than the first corrugation initial area or the second corrugation initial area. Similarly, multiple groups of second detection result data based on the second detection feature and multiple groups of third detection result data based on the third detection feature can respectively determine the corrugation areas obtained by their respective fusion.

[0081] The number of occurrences of each corrugation level in the corrugation area is counted based on the corrugation area determined based on each detection feature, and the confidence of each corrugation level in the corrugation area is calculated based on the number of occurrences of each corrugation level in the corrugation area.

[0082] In this embodiment, data fusion is performed based on multiple groups of detection result data corresponding to the same detection feature to jointly determine the corrugation area where corrugation occurs, and the number of occurrences of each corrugation level in the corrugation area is counted to ensure the accuracy of the subsequent calculation of the confidence level of each corrugation level.

[0083] Among them, the calculation expression of the confidence level of each corrugation level is:

[0084]

[0085] Among them, p(y,x) is the confidence of corrugation level x in area range y, m is the number of times corrugation level x appears in area range y, and M is the total number of times all corrugation levels appear in area range y.

[0086] Embodiment 4

[0087] On the basis of the above embodiment, the step of merging the multiple groups of initial corrugation areas of the same corrugation mill to obtain the corrugation area comprises:

[0088] Detecting whether the multiple groups of initial areas of the same corrugation mill overlap;

[0089] When the multiple groups of corrugation initial areas of the same corrugation mill overlap, the sum of the corrugation initial areas is determined as the corrugation area.

[0090] In this embodiment, the corrugation initial area of ​​each corrugation mill is determined respectively based on multiple groups of detection result data corresponding to the same detection feature, and it is detected whether at least two corrugation initial areas of the same corrugation mill overlap. When at least two corrugation initial areas of the same corrugation mill overlap, the sum of the corrugation initial areas is determined to be the corrugation area.

[0091] In one embodiment, the number of groups of detection result data corresponding to each detection feature is multiple groups, and multiple groups of corrugation defect tables are respectively determined based on the multiple groups of detection result data corresponding to the same detection feature, each corrugation defect table includes the corrugation initial area where each corrugation appears and the number of times each corrugation level appears in the corrugation initial area; the corrugation defect table determined by one group of detection result data is defined as the basic corrugation defect table, each corrugation initial area in the basic corrugation defect table is defined as the basic corrugation initial area, and the corrugation initial area determined by the remaining groups of detection result data is the calibration corrugation initial area, and each corrugation initial area in the calibration corrugation initial area is the calibration corrugation initial area; it is detected whether the calibration corrugation initial area of ​​the same corrugation mill overlaps with the basic corrugation initial area, and when the calibration corrugation initial area of ​​the same corrugation mill overlaps with the basic corrugation initial area, the corrugation area of ​​the corrugation mill is the sum of the calibration corrugation initial area and the basic corrugation initial area.

[0092] Embodiment 5

[0093] On the basis of the above embodiment, the step of merging the multiple groups of initial corrugation areas of the same corrugation mill to obtain the corrugation area further comprises:

[0094] Detecting whether the difference between the starting positions of the plurality of groups of initial areas of the same corrugation mill is within a preset range;

[0095] When the difference between the starting positions of a plurality of groups of the corrugation initial regions of the same corrugation mill is within a preset range, each of the corrugation initial regions is determined as the corrugation region.

[0096] In this embodiment, the corrugation initial area of ​​each corrugation mill is determined respectively based on multiple groups of detection result data corresponding to the same detection feature, and it is detected whether the difference between the starting positions of at least two corrugation initial areas of the same corrugation mill is within a preset range. When the difference between the starting positions of at least two corrugation initial areas of the same corrugation mill is within the preset range, each of the corrugation initial areas is determined to be the corrugation area.

[0097] In one embodiment, the number of groups of detection result data corresponding to each detection feature is multiple groups, and multiple groups of corrugation defect tables are respectively determined based on the multiple groups of detection result data corresponding to the same detection feature, each corrugation defect table includes the corrugation initial area where each corrugation appears and the number of times each corrugation level appears in the corrugation initial area; the corrugation defect table determined by one group of detection result data is defined as the basic corrugation defect table, each corrugation initial area in the basic corrugation defect table is defined as the basic corrugation initial area, and the corrugation initial area determined by the remaining groups of detection result data is the calibration corrugation initial area, and each corrugation initial area in the calibration corrugation initial area is the calibration corrugation initial area; it is detected whether there is a corrugation in the calibration corrugation initial area whose starting position is the same as the starting position of the same corrugation in the basic corrugation initial area, and when there is a corrugation in the calibration corrugation initial area whose starting position is the same as the starting position of the same corrugation in the basic corrugation initial area, the corrugation area of ​​the corrugation is the basic corrugation initial area.

[0098] Embodiment 6

[0099] On the basis of the above embodiment, the step of fusing the confidences of the same corrugation level determined by at least two detection features of the same corrugation area to obtain the fused confidence of the same corrugation level comprises:

[0100] When the number of the detection features is two, data fusion is performed on the confidences of the same corrugation level corresponding to the two detection features of the same corrugation area to obtain the fused confidence of the same corrugation level;

[0101] When the data of the detection feature is N, the confidences of the same corrugation level corresponding to two detection features of the same corrugation area are data fused to obtain a fusion result, and the fusion result is data fused with the confidence of the same corrugation level corresponding to another detection feature, until the data of the confidences of the same corrugation level corresponding to N detection features are fused to obtain the fused confidences of the same corrugation level corresponding to all the detection features; N is an integer greater than or equal to 3.

[0102] In this embodiment, the corrugation area is determined based on the detection result data corresponding to each detection feature, and the confidence of each corrugation level of each corrugation area is calculated. Then, the confidence of the same corrugation level corresponding to different detection features of the same corrugation area is respectively processed to obtain the fusion confidence of the same corrugation level.

[0103] When the number of detection features is two, the confidences of the same corrugation level of the two detection features of the same corrugation area are data fused to obtain the fused confidence of the same corrugation level; for example, there are two detection features, namely the first detection feature and the second detection feature, and the confidences of each corrugation level in each corrugation area can be obtained based on the detection result data of each detection feature. For example, the corrugation area includes areas 1, 2 and 3, wherein each corrugation level in each corrugation area has a corresponding confidence, and the corrugation level includes A, B and C. Then, based on the first detection feature, the confidence corresponding to the corrugation level A in area 1 is determined, based on the first detection feature, the confidence corresponding to the corrugation level B in area 1 is determined; based on the first detection feature, the confidence corresponding to the corrugation level A in area 2 is determined. Similarly, based on multiple groups of detection result data of the second detection feature, each corrugation area is obtained, wherein each corrugation level in each corrugation area has a corresponding confidence; based on the second detection feature, the confidence corresponding to the corrugation level A in area 1 is determined. The confidences of the same corrugation level of the same corrugation area obtained based on the two detection features are fused. For example, the confidence of the corrugation level A of area 1 obtained based on the first detection feature is fused with the confidence of the corrugation level A of area 1 obtained based on the second detection feature to obtain the fused confidence of the corrugation level A of area 1. Similarly, the fused confidence of the corrugation level B of area 1 and the fused confidence of the corrugation level C of area 1 can be obtained. Subsequently, the occurrence of corrugation in area 1 is detected based on the fused confidences of the above corrugation levels, which can ensure the accuracy of the detection.

[0104] When the number of detection features is N, and N is an integer greater than or equal to 3, data fusion is performed on the confidences of the same corrugation level of two detection features in the same corrugation area to obtain a fusion result, i.e., the fusion confidence of the same corrugation level, and the fusion result is data fused with the confidence of the same corrugation level of another detection feature until the data fusion of the confidences of the same corrugation level corresponding to N detection features is obtained to obtain the fusion confidence of the same corrugation level corresponding to all detection features; for example, there are three detection features, namely the first detection feature, the second detection feature, and the third detection feature, and the confidence of each corrugation level in each corrugation area can be obtained based on the detection result data of each detection feature, wherein the corrugation area includes areas 1, 2, and 3, wherein each corrugation level in each corrugation area has a corresponding confidence, The corrugation levels include A, B and C for explanation. First, the confidence level of corrugation level A for area 1 determined based on the first detection feature and the confidence level of corrugation level A for area 1 determined based on the second detection feature are fused to obtain the fused confidence level of corrugation level A for area 1, i.e., the fusion result. The prior fused confidence level is fused with the confidence level of corrugation level A for area 1 determined based on the third detection feature to obtain the fused confidence level. By analogy, the data fusion of all confidence levels of corrugation level A of area 1 with different detection features is completed. Similarly, the fused confidence level of corrugation level B for area 1 and the fused confidence level of corrugation level C for area 1 can be obtained. Subsequently, the occurrence of corrugation in area 1 is detected based on the fused confidence levels of the above-mentioned corrugation levels to ensure the accuracy of detection.

[0105] Embodiment 7

[0106] On the basis of the above embodiment, the step of fusing the confidences of the same corrugation level determined by at least two detection features of the same corrugation area to obtain the fused confidence of the same corrugation level comprises:

[0107] Based on the evidence theory, data fusion is performed on the confidence of the same corrugation level determined by at least two detection features of the same corrugation area to obtain the fused confidence of the same corrugation level.

[0108] In this embodiment, the DS (Dempster-Shafer) evidence theory algorithm is used to perform data fusion on the confidence of the same corrugation level of at least two detection features in the same corrugation area. The evidence theory algorithm regards each piece of collected perception data as evidence to support the reasoning conclusion in the form of reasoning, and obtains the fusion result by integrating various data evidence. According to its fusion characteristics, it can be seen that the DS evidence theory is more suitable for decision-making-level data fusion, and can make accurate decisions based on the collected perception data. The normalization coefficient calculation formula of the DS evidence theory algorithm is as follows:

[0109]

[0110] In the formula, m1(B i ) is the first device B that needs to be integrated i Confidence level of the level, m2(C j ) is the C of the second device to be integrated i The confidence level of the level, m1(B i ) and m2(C j ) can be multiplied if: B i With C j The intersection of does not include the focal element "A".

[0111] In the framework of evidence theory, the combined mass functions of the focal element “A” can be fused through the Dempster combination rule:

[0112]

[0113] Where K is the normalization coefficient of "A", m1(B i ) and m2(C j ) can be multiplied if: B i With C j The intersection of includes and only includes focal element "A". Similarly, the combined mass function of focal element "B" and focal element "C" is:

[0114]

[0115]

[0116] Embodiment 8

[0117] Based on the above embodiments, this embodiment provides an application example. Figures 2 to 4 shown.

[0118] like Figure 2 As shown, the rail corrugation detection method comprises:

[0119] Step 1: Collect monitoring data.

[0120] The collecting of monitoring data includes obtaining detection result data corresponding to at least two detection features for characterizing rail corrugation.

[0121] There are three detection features used to characterize rail corrugation, namely, the axle box lateral acceleration signal (ie, the axle box relative acceleration), the wheel-rail force signal, and the track acoustic signal.

[0122] The test result data includes the starting and ending positions of the corrugation, as well as the number of times the corrugation level occurs here.

[0123] In this embodiment, before collecting monitoring data, it also includes installing a sensor device for acquiring detection features.

[0124] In one embodiment, a vibration acceleration sensor may be installed on the train to obtain a vibration acceleration signal, and the detection result data is obtained by analyzing the vibration acceleration signal, that is, the starting position of the corrugation is determined and the number of occurrences of the corrugation level at this location is counted.

[0125] In one embodiment, a force sensor is installed on the wheel rail of the train to obtain a wheel rail force signal, and the wheel rail force signal of the force sensor is analyzed to obtain the detection result data, that is, to determine the starting position of the corrugation and count the number of times the corrugation level occurs there.

[0126] In one embodiment, microphones are installed on the wheels and rails of the train to obtain track acoustic signals, and the track acoustic signals are analyzed. Specifically, through the wideband adaptive analysis of the algorithm, the starting position of the rail corrugation is extracted and the number of times the corrugation occurs here is counted.

[0127] In one embodiment, data is acquired through three trains, each of which is equipped with an axial vibration acceleration sensor, and two of which are equipped with sensors related to wheel-rail force and track acoustics. At the same time, the above-mentioned detection result data, that is, corrugation disease, are written into the position signal disease table for subsequent query of the disease location at any time.

[0128] Step 2: In-device data fusion, calculate the confidence of each device.

[0129] Data fusion within the device and calculation of the confidence level of each device include determining the detection results of a single device and determining the confidence level based on the probability of occurrence of the detection results.

[0130] The data fusion within the equipment and the calculation of the confidence of each equipment include two aspects: creating a position information table (mainly to determine whether the diagnosed corrugation position belongs to the same corrugation) and solving different levels of confidence.

[0131] A device (i.e., a sensor for obtaining detection features) is installed on a train. When the train is operating on the same section of rail line, it will repeatedly travel on this line. After analyzing the collected data, it is found that the detection result data of a single device may have problems of inaccurate detection, misjudgment or missed detection. Therefore, it is necessary to statistically analyze the results of multiple diagnoses, including the diagnosis position and diagnosis level. The diagnosis position and diagnosis level are stored in the location information table, which stores the diagnosis position and level.

[0132] like Figure 3 As shown, first, taking the wave grinding positioning of the first train as an example, a disease table containing the wave grinding position and grade is created, and then the information of other trains is traversed to determine whether the starting position information of the wave grinding diagnosed by the train exists in the disease table; if it exists, the left and right standards of the first wave grinding disease information are set as the benchmark, and the subsequent wave grinding disease information is traversed to compare and determine the end position of the wave grinding disease. If the end position is within a threshold range compared with the end position in the disease table, it means that the detected disease and the one in the disease table belong to the same disease, otherwise, new position information and wave grinding grade are added to the information table; if it does not exist, first set the position information, that is, the starting position of the wave grinding disease, to d, and then determine whether the difference dc between d and the existing position c in the position information table is less than the threshold; if the difference is less than the threshold, take the position value that makes dc the smallest and attribute the disease to this position, and if the difference is greater than the threshold, also add new position information and wave grinding grade to the information table.

[0133] After completing the above steps, count the number of occurrences of the corresponding disease level at a location, and combine the number of disease occurrences to obtain the confidence level. The confidence level calculation formula is:

[0134]

[0135] In the formula, p(1,A) is the confidence level of corrugation grade A at position 1, m is the number of occurrences of corrugation grade A at position 1, and M is the total number of occurrences of all grades of all corrugation diseases at all positions. At this point, data fusion within the device is completed.

[0136] Step 3: Data fusion between devices.

[0137] Data fusion between devices includes adding expert opinions on the basis of single device confidence according to DS evidence theory, and realizing data interoperability and fusion of multi-source data by integrating the levels of rail corrugation diseases detected by three technologies: axle box vibration acceleration, wheel-rail force and track acoustics. Ultimately, a decision on rail corrugation diseases is made, and the decision result is the probability of the presence of corrugation diseases in a certain section.

[0138] In this embodiment, there are three trains that can detect rail corrugation diseases. Each of the three trains is equipped with an axial vibration acceleration sensor, and two of the trains are equipped with sensors related to wheel-rail force and track acoustics.

[0139] Among them, the idea of ​​data fusion between devices is: if there is only axial vibration acceleration data, this data will be used as the judgment standard; if there are only axial vibration acceleration data and wheel-rail force data, the data of these two devices will be fused; if there are only axial vibration acceleration data and track acoustic data, the data of these two devices will be fused; if data of all three devices exist, all data will be fused. For example, first, the detection result data corresponding to the axial lateral acceleration signal and the detection result data corresponding to the wheel-rail force signal are fused to obtain a new set of data, and then the new data is fused with the track acoustic signal for a second time between the devices. The final fusion result is the data fusion result of the three devices.

[0140] It is worth noting that among the three devices, any two devices can be fused with DS evidence theory first and then fused with the remaining one, which will have no effect on the final result.

[0141] like Figure 4 As shown, the specific process of inter-device fusion is as follows: first, the detection result data corresponding to the axial lateral acceleration signal and the detection result data corresponding to the wheel-rail force signal are fused to obtain a new set of data, and then the new data is subjected to secondary inter-device data fusion with the track acoustic signal.

[0142] The corrugation diagnosis results (ie, test result data) corresponding to the axle box vibration acceleration, wheel-rail force and track acoustics within one week are extracted. The corrugation diagnosis results include two aspects: position and level.

[0143] On the one hand, the results of axial vibration acceleration are fused and the confidence is calculated as a prerequisite for data fusion between devices. On the other hand, it is determined whether there is any critical information. If there is, the critical information is directly sent to the terminal for early warning; if there is no critical information, first determine whether there is diagnostic data of wheel-rail force and / or track acoustics. If so, create a corresponding location information table. First, the data of wheel-rail force and axle box vibration acceleration are fused, and then the fused data results are fused with the track acoustic data to obtain the decision-level fusion result. If there is no data related to wheel-rail force and track acoustics, the detection result of axle box vibration acceleration is directly used as the final result.

[0144] The algorithm used for data fusion between devices is the DS (Dempster-Shafer) evidence theory algorithm. This data fusion method uses reasoning to regard each piece of sensory data collected as evidence to support the reasoning conclusion, and integrates various data evidence to obtain the fusion result. According to its fusion characteristics, it can be seen that the DS evidence theory is more suitable for decision-making-level data fusion and can make accurate decisions based on the collected sensory data. The normalization coefficient calculation formula of the DS evidence theory algorithm is as follows:

[0145]

[0146] In the formula, m1(B i ) is the first device B that needs to be integrated i Confidence level of the level, m2(C j ) is the C of the second device to be integrated i The confidence level of the level, m1(B i ) and m2(C j ) can be multiplied if: B i With C j The intersection of does not include the focal element "A".

[0147] In the framework of evidence theory, the combined mass functions of the focal element “A” can be fused through the Dempster combination rule:

[0148]

[0149] Where K is the normalization coefficient of "A", m1(B i ) and m2(C j ) can be multiplied if: B i With C j The intersection of includes and only includes focal element "A". Similarly, the combined mass function of focal element "B" and focal element "C" is:

[0150]

[0151] Assuming that the rail corrugation diseases identified by the three devices are E1 (axle box vibration acceleration), E2 (wheel-rail force), and E3 (track acoustics), the focal elements are A (wave depth < 0.3), B (0.3 < wave depth < 0.5), and C (wave depth > 0.5) (non-intersecting), as shown in Table 1, which is the result of the fusion within the three devices. The table gives in detail the start-end position of the diagnosis of each device, as well as the number and confidence of the diagnosis level.

[0152] Table 1 Fusion results in each device

[0153]

[0154]

[0155] Taking the confidence of three devices with a starting position of about 90 and an ending position of about 120 in Table 1 as an example, we perform data fusion calculation between devices. First, we fuse the two devices E1 and E2, and calculate the normalization coefficients of E1 and E2:

[0156] K 1,2 =1-(m1(A)×m2(B)+m1(A)×m2(C)+m1(B)×m2(A)

[0157] +m1(B)×m2(C)+m1(B)×m2(A)+m1(C)×m2(B))

[0158] =1-(0.72×0.6+0.72×0+0.28×0.4+0.28×0+0×0.4+0×0.6)

[0159] =1-0.544

[0160] =0.456

[0161] The probability of fusion of E1 and E2 is calculated as:

[0162]

[0163] Then, the result of the fusion of the two devices E1 and E2 (recorded as: E12) is fused with the E3 device, and the normalization coefficient of E12 and E3 is calculated as follows:

[0164] K 12,3 =1-(m 12 (A)×m2(B)+m 12 (A)×m2(C)+m 12 (B)×m2(A)

[0165] +m 12 (B)×m2(C)+m 12 (B)×m2(A)+m 12 (C)×m2(B))

[0166] =1-(0.632×0.33+0.632×0+0.368×0.67+0.368×0+0×0+0×0)

[0167] =1-0.455

[0168] =0.545

[0169] The probability calculation after the fusion of E12 and E3 devices is:

[0170]

[0171] Similarly, data fusion is performed on the corrugation diagnosed by the three devices at all starting and ending positions. The fused results are shown in Table 2.

[0172] Table 2. Data fusion results between devices

[0173]

[0174] In this way, the fused confidence of each corrugation level of each corrugation area where corrugation occurs is obtained, and the occurrence of corrugation in the rail section is detected based on the fused confidence of each corrugation level of each corrugation area.

[0175] In this embodiment, firstly, the clustering method is used to fuse the wave grinding positions and levels diagnosed by single equipment detection, that is, the wave grindings with similar positions are classified and counted; thus, the accuracy of the single equipment detection results is improved; secondly, the data obtained by the three equipment detections of axle box vibration acceleration, wheel-rail force and track acoustics are directly sent to a server port and stored in the big data platform, thus realizing the unified management and call of all data and making data reading more convenient; finally, in this embodiment, the DS evidence theory algorithm is used to fuse the multi-source data detection results (wave grinding detected by axle box vibration acceleration, wheel-rail force and track acoustics, including wave grinding positions and levels), so as to save operation and maintenance costs and reduce the operation risks of the track system.

[0176] Embodiment 9

[0177] Based on the above embodiment, this embodiment provides a computer device, whose internal structure diagram can be as follows: Figure 5 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program, and the non-volatile storage medium is deployed with a database, which is used to store template images. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with other computer devices deployed with application software. When the computer program is executed by the processor, a rail corrugation detection method is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a key, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse, etc.

[0178] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0179] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the following steps:

[0180] Acquire at least two detection result data corresponding to the detection features for characterizing rail corrugation; wherein each detection feature corresponds to one detection result data;

[0181] Determine the corrugation area where corrugation occurs based on each of the detection result data, and calculate the confidence of each corrugation level in the corrugation area; the confidence is used to characterize the probability of occurrence of the corrugation level;

[0182] Performing data fusion on the confidences of the same corrugation level determined by at least two detection features of the same corrugation area to obtain fused confidence of the same corrugation level;

[0183] The occurrence of corrugation in the rail section is detected based on the fused confidence of each corrugation level in each corrugation area.

[0184] In one embodiment, when the processor executes the computer program, the following steps are implemented:

[0185] The detection result data includes the starting position and the ending position of each corrugation in the rail section and the number of times each corrugation level occurs;

[0186] Determine the initial corrugation area of ​​each corrugation according to the starting position and the ending position of each corrugation in the detection result data;

[0187] Based on at least two groups of detection result data corresponding to the same detection feature, multiple groups of corrugation initial areas are obtained;

[0188] Merging the multiple groups of initial corrugation areas of the same corrugation mill to obtain a corrugation mill area;

[0189] Counting the number of occurrences of each corrugation level in the corrugation area;

[0190] The confidence level of each corrugation level in each corrugation area is calculated based on the number of occurrences of each corrugation level in each corrugation area.

[0191] In one embodiment, when the processor executes the computer program, the following steps are implemented:

[0192] Detecting whether the multiple groups of initial areas of the same corrugation mill overlap;

[0193] When the multiple groups of corrugation initial areas of the same corrugation mill overlap, the sum of the corrugation initial areas is determined as the corrugation area.

[0194] In one embodiment, when the processor executes the computer program, the following steps are implemented:

[0195] Detecting whether the difference between the starting positions of the plurality of groups of initial areas of the same corrugation mill is within a preset range;

[0196] When the difference between the starting positions of a plurality of groups of the corrugation initial regions of the same corrugation mill is within a preset range, each of the corrugation initial regions is determined as the corrugation region.

[0197] In one embodiment, when the processor executes the computer program, the following steps are implemented:

[0198] When the number of the detection features is two, data fusion is performed on the confidences of the same corrugation level corresponding to the two detection features of the same corrugation area to obtain the fused confidence of the same corrugation level;

[0199] When the data of the detection feature is N, and N is an integer greater than or equal to 3, data fusion is performed on the confidences of the same corrugation level corresponding to two detection features of the same corrugation area to obtain a fusion result, and the fusion result is data fused with the confidence of the same corrugation level corresponding to another detection feature, until the data of the confidences of the same corrugation level corresponding to N detection features are fused to obtain the fused confidences of the same corrugation level corresponding to all the detection features.

[0200] In one embodiment, when the processor executes the computer program, the following steps are implemented:

[0201] Based on the evidence theory, data fusion is performed on the confidence of the same corrugation level determined by at least two detection features of the same corrugation area to obtain the fused confidence of the same corrugation level.

[0202] In some implementations of this embodiment, a computer program product is provided, including a computer program / instructions, characterized in that when the computer program is executed by a processor, the steps of the method described in the above embodiment are implemented.

[0203] The processor may include, but is not limited to, one or more processors or microprocessors. Each processor may be an application specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field programmable gate array (FPGA), a controller, a microcontroller, a microprocessor or other electronic components to execute the method in the above embodiments.

[0204] The computer-readable storage medium may be implemented by any type of volatile or non-volatile storage device or a combination thereof, and the computer-readable storage medium may include but is not limited to, for example, random access memory (RAM), read-only memory (ROM), flash memory, EPROM memory, EEPROM memory, registers, computer storage media (e.g., hard disk, floppy disk, solid-state drive, removable disk, CD-ROM, DVD-ROM, Blu-ray disc, etc.).

[0205] The computer-readable storage medium may also store at least one computer executable program / instruction, which may be, for example, a computer-readable instruction. The computer-readable storage medium includes, but is not limited to, for example, a volatile memory and / or a non-volatile memory. The volatile memory may include, for example, a random access memory (RAM) and / or a cache memory (cache), etc. The computer-readable storage medium may include, for example, a read-only memory (ROM), a hard disk, a flash memory, etc. For example, a non-transitory computer-readable storage medium may be connected to a computing device such as a computer, and then, when the computing device runs the computer-readable instructions stored on the computer-readable storage medium, the various methods described above may be performed.

[0206] In addition, the computer device may also include (but not limited to) a data bus, an input / output (I / O) bus, a display, and input / output devices (eg, keyboard, mouse, speaker, etc.), etc.

[0207] The processor may communicate with external devices via an I / O bus via a wired or wireless network.

[0208] In one embodiment, the at least one computer executable instruction may also be compiled into or constitute a software product / computer program product, wherein one or more computer executable instructions are executed by a processor to perform the various functions and / or method steps in the embodiments described in the present technology.

[0209] In the embodiments provided in the present disclosure, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the above-mentioned module, program segment or a part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0210] It should be noted that in the present disclosure, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element limited by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0211] Although the embodiments disclosed in the present disclosure are as above, the above contents are only embodiments adopted for facilitating the understanding of the present disclosure and are not intended to limit the present disclosure. Any technician in the technical field to which the present disclosure belongs can make any modifications and changes in the form and details of the implementation without departing from the spirit and scope disclosed in the present disclosure, but the scope of patent protection of the present disclosure shall still be subject to the scope defined in the attached claims.

Claims

1. A rail corrugation detection method, characterized in that: include: Acquire at least two detection result data corresponding to the detection features for characterizing rail corrugation; wherein each detection feature corresponds to one detection result data; Determine the corrugation area where corrugation occurs based on each of the detection result data, and calculate the confidence of each corrugation level in the corrugation area; the confidence is used to characterize the probability of occurrence of the corrugation level; Performing data fusion on the confidences of the same corrugation level determined by at least two detection features of the same corrugation area to obtain fused confidence of the same corrugation level; The occurrence of corrugation in the rail section is detected based on the fused confidence of each corrugation level in each corrugation area.

2. The rail corrugation detection method according to claim 1, characterized in that: The detection features used to characterize rail corrugation include axle box vibration acceleration, wheel-rail force and track acoustics.

3. The rail corrugation detection method according to claim 1, characterized in that: The detection result data includes the starting position and the ending position of each corrugation in the rail section and the number of times each corrugation level occurs; The step of respectively determining the corrugation areas where corrugation occurs based on the detection result data and calculating the confidence of each corrugation level of the corrugation area comprises: Determine the initial corrugation area of ​​each corrugation according to the starting position and the ending position of each corrugation in the detection result data; Based on at least two groups of detection result data corresponding to the same detection feature, multiple groups of corrugation initial areas are obtained; Merging the multiple groups of initial corrugation areas of the same corrugation mill to obtain a corrugation mill area; Counting the number of occurrences of each corrugation level in the corrugation area; The confidence level of each corrugation level in each corrugation area is calculated based on the number of occurrences of each corrugation level in each corrugation area.

4. The rail corrugation detection method according to claim 3, characterized in that: The step of merging the multiple groups of initial corrugation areas of the same corrugation mill to obtain the corrugation area comprises: Detecting whether the multiple groups of initial areas of the same corrugation mill overlap; When the multiple groups of corrugation initial areas of the same corrugation mill overlap, the sum of the corrugation initial areas is determined as the corrugation area.

5. The rail corrugation detection method according to claim 4, characterized in that: The step of merging the multiple groups of initial corrugation areas of the same corrugation mill to obtain the corrugation area further comprises: Detecting whether the difference between the starting positions of the plurality of groups of initial areas of the same corrugation mill is within a preset range; When the difference between the starting positions of a plurality of groups of the corrugation initial regions of the same corrugation mill is within a preset range, each of the corrugation initial regions is determined as the corrugation region.

6. The rail corrugation detection method according to claim 1, characterized in that: The step of fusing the confidences of the same corrugation level determined by at least two detection features of the same corrugation area to obtain the fused confidence of the same corrugation level comprises: When the number of the detection features is two, data fusion is performed on the confidences of the same corrugation level corresponding to the two detection features of the same corrugation area to obtain the fused confidence of the same corrugation level; When the data of the detection feature is N, and N is an integer greater than or equal to 3, data fusion is performed on the confidences of the same corrugation level corresponding to two detection features of the same corrugation area to obtain a fusion result, and the fusion result is data fused with the confidence of the same corrugation level corresponding to another detection feature, until the data of the confidences of the same corrugation level corresponding to N detection features are fused to obtain the fused confidences of the same corrugation level corresponding to all the detection features.

7. The rail corrugation detection method according to claim 1, characterized in that: The step of fusing the confidences of the same corrugation level determined by at least two detection features of the same corrugation area to obtain the fused confidence of the same corrugation level comprises: Based on the evidence theory, data fusion is performed on the confidence of the same corrugation level determined by at least two detection features of the same corrugation area to obtain the fused confidence of the same corrugation level.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images