Wheel Circumference Surface Detection Method, Device, Equipment and Storage Medium for Rail Vehicles

Abstract

The present invention discloses a method, device, equipment and computer-readable storage medium for detecting the circumferential surface of a wheel of a rail vehicle. The method includes collecting current fluctuation information of the undisturbed circumferential surface undulation of the wheel; determining the theoretical vibration data of the wheel according to the current fluctuation information and a pre-created correspondence model between the fluctuation information and the vibration data of the wheel; finding a set of historical vibration data segments with the highest similarity to the theoretical vibration data in the historical vibration data, and using the vibration data segments adjacent to the historical vibration data segments as the estimated vibration data of the wheel; determining the corresponding estimated fluctuation information according to the estimated vibration data, and combining the current fluctuation information to determine the fluctuation information of the complete circumferential surface of the wheel. Under the condition that only the current fluctuation information of a part of the wheel is collected in this application, the waveform information of the undetected circumferential surface of the wheel is determined, reducing the detection difficulty of the circumferential surface information of the wheel.

Description

Technical Field

[0001] The present invention relates to the field of railway vehicle calculations, and particularly to a method, device, equipment, and computer-readable storage medium for detecting the circumferential surface of the wheels of a railway vehicle. Background Art

[0002] During the actual operation of the wheels of a railway vehicle, wear will occur as the operation time extends, and thus the circumferential surface of the wheels will present a polygonal surface due to uneven wear. The polygonal surface of the wheels will generate abnormal high-frequency vibrations between the wheel and the rail, which will have an important impact on the reliability of railway vehicle components and train operation safety. Therefore, in vehicle operation and maintenance, regularly measuring the polygonal loss of the wheels is crucial for the safe and sound operation of railway vehicles.

[0003] During the process of measuring the polygonal loss of the circumferential surface of the wheels, it is necessary to slide the measuring device to cover the entire circumferential surface of the wheels. However, the actual structure of the railway vehicle determines that there will inevitably be a part of the area on the wheels that is blocked. Therefore, currently, in order to measure the entire circumferential surface of the wheels, it is necessary to relieve the vehicle braking, use equipment such as a jack to lift the vehicle to separate the wheels from the rails, and rotate the wheels to achieve the measurement of the entire circumferential surface of the wheels.

[0004] This method not only requires the assistance of more tooling, resulting in high costs, but also the entire testing process is rather cumbersome, time-consuming, and even requires the disassembly of the installation components associated with the wheel set. Summary of the Invention

[0005] The purpose of the present invention is to provide a method, device, equipment, and computer-readable storage medium for detecting the circumferential surface of the wheels of a railway vehicle, which simplifies the difficulty of detecting the circumferential surface of the wheels to a certain extent.

[0006] To solve the above technical problems, the present invention provides a method for detecting the circumferential surface of the wheels of a railway vehicle, including:

[0007] Collecting the current fluctuation information of the undisturbed circumferential surface undulation of the wheels;

[0008] Determining the theoretical vibration data of the wheels according to the current fluctuation information and the correspondence relationship model between the fluctuation information and the vibration data of the wheels created in advance;

[0009] Searching for a set of historical vibration data segments with the highest similarity to the theoretical vibration data in the historical vibration data, and using the vibration data segments adjacent to the historical vibration data segments in the historical vibration data as the estimated vibration data of the wheels;

[0010] Determine the estimated fluctuation information corresponding to the circumferential surface of the wheel except for the circumferential surface corresponding to the current fluctuation information according to the estimated vibration data, and determine the fluctuation information of the complete circumferential surface of the wheel according to the estimated fluctuation information and the current fluctuation information.

[0011] Preferably, the process of pre-creating the correspondence model includes:

[0012] Collect vibration data samples during the operation of the wheel and fluctuation information samples of the entire circumferential surface of the wheel;

[0013] Perform neural network training according to the vibration data samples and fluctuation information samples respectively corresponding to each position point on the wheel when it rotates to fit the track, and obtain the correspondence model.

[0014] Preferably, taking the vibration data section adjacent to the historical vibration data section as the estimated vibration data of the wheel in the historical vibration data includes:

[0015] Taking a vibration data section with a set section length adjacent to the historical vibration data section in the historical vibration data as the estimated vibration data; wherein, the set section length is determined according to the wheel speed corresponding to the collection of the historical vibration data.

[0016] Preferably, determining the estimated fluctuation information corresponding to the circumferential surface of the wheel except for the circumferential surface corresponding to the current fluctuation information according to the estimated vibration data includes:

[0017] Obtain the estimated fluctuation information of the occluded circumferential surface of the wheel according to the estimated vibration data and the correspondence model.

[0018] Preferably, finding a set of historical vibration data sections with the highest similarity to the theoretical vibration data in the historical vibration data includes:

[0019] Compare the historical fluctuation information corresponding to the historical vibration data with the current fluctuation information of the wheel, and find multiple historical fluctuation information sections with a similarity not lower than the first set similarity to the current fluctuation information;

[0020] Compare the similarity between the historical vibration data sections in the historical vibration data corresponding to each of the historical fluctuation information sections and the theoretical vibration data, and determine a set of historical vibration data sections with the highest similarity to the theoretical vibration data.

[0021] A detection device for the circumferential surface of a wheel of a rail vehicle, comprising:

[0022] A data acquisition module for acquiring the current fluctuation information of the undisturbed circumferential surface undulation of the wheel;

[0023] A first operation module for determining the theoretical vibration data of the wheel according to the current fluctuation information and a pre-created correspondence model between the fluctuation information and the vibration data of the wheel;

[0024] A second operation module for finding a set of historical vibration data segments with the highest similarity to the theoretical vibration data in the historical vibration data, and using the vibration data segments adjacent to the historical vibration data segments in the historical vibration data as the predicted vibration data of the wheel;

[0025] A third operation module for determining the predicted fluctuation information corresponding to the circumferential surface of the wheel except the circumferential surface corresponding to the current fluctuation information according to the predicted vibration data, and determining the fluctuation information of the complete circumferential surface of the wheel according to the predicted fluctuation information and the current fluctuation information.

[0026] Preferably, it further includes a model creation module for acquiring vibration data samples during the operation of the wheel and fluctuation information samples of the entire circumferential surface of the wheel; performing neural network training according to the vibration data samples and fluctuation information samples respectively corresponding to each position point on the wheel when rotating to fit the track, and obtaining the correspondence model.

[0027] Preferably, the second operation module is specifically configured to compare the historical fluctuation information corresponding to the historical vibration data with the current fluctuation information of the wheel, and find multiple historical fluctuation information segments with a similarity not lower than a first set similarity to the current fluctuation information; compare the similarity between the historical vibration data segments in the historical vibration data respectively corresponding to each segment of the historical fluctuation information segments and the theoretical vibration data, and determine a set of historical vibration data segments with the highest similarity to the theoretical vibration data.

[0028] A detection device for the circumferential surface of a wheel of a rail vehicle, comprising:

[0029] A memory for storing a computer program;

[0030] A processor for executing the computer program to implement the steps of the detection method for the circumferential surface of the wheel of a rail vehicle as described in any one of the above.

[0031] A computer-readable storage medium storing a computer program, and the computer program is executed to implement the steps of the detection method for the circumferential surface of the wheel of a rail vehicle as described in any one of the above.

[0032] The detection method, device, equipment and computer-readable storage medium for the circumferential surface of the wheel of a rail vehicle provided by the present invention. The method includes collecting the current fluctuation information of the undisturbed circumferential surface undulation of the wheel; determining the theoretical vibration data of the wheel according to the current fluctuation information and the pre-created correspondence model between the fluctuation information and the vibration data of the wheel; finding a set of historical vibration data segments with the highest similarity to the theoretical vibration data in the historical vibration data, and using the vibration data segments adjacent to the historical vibration data segments in the historical vibration data as the estimated vibration data of the wheel; determining the estimated fluctuation information corresponding to the circumferential surface of the wheel except the circumferential surface corresponding to the current fluctuation information according to the estimated vibration data, and determining the fluctuation information of the complete circumferential surface of the wheel according to the estimated fluctuation information and the current fluctuation information.

[0033] In this application, by using the characteristic that there is a corresponding correlation between the undulation of the circumferential surface of the wheel and the vibration data during the operation of the wheel, under the condition of only collecting the current fluctuation information of a part of the circumferential surface of the wheel, through the correlation model between the fluctuation information and the vibration data, the theoretical vibration data corresponding to the current fluctuation information is determined, and then the estimated vibration data closest to the circumferential surface of the wheel is determined in the historical vibration data by using the theoretical vibration data, and the estimated vibration data is used as the vibration data of the undetected circumferential surface of the wheel. Based on this estimated vibration data, the waveform information of the undetected circumferential surface of the wheel can be determined, and combined with the currently measured current fluctuation information, the fluctuation information of the complete circumferential surface of the wheel can be obtained, and this fluctuation information is the polygon information of the wheel. During the whole detection process, no additional tools such as jacks are needed, reducing the consumption of manpower and material resources and reducing the detection difficulty of the circumferential surface information of the wheel. Brief Description of the Drawings

[0034] In order to more clearly illustrate the technical solutions of the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained according to these drawings.

[0035] Figure 1 It is a schematic flowchart of the detection method for the circumferential surface of the wheel of a rail vehicle provided by an embodiment of the present application;

[0036] Figure 2 It is a schematic structural diagram of the measuring device for measuring the fluctuation information of the wheel provided by this embodiment;

[0037] Figure 3 For Figure 2 It is a schematic cross-sectional view of the contact part between the measuring device and the wheel in

[0038] Figure 4 It is a structural block diagram of a detection device for the circumferential surface of a wheel of a rail vehicle provided by an embodiment of the present invention;

[0039] Figure 5 It is a structural block diagram of a detection device for the circumferential surface of a wheel of a rail vehicle provided by an embodiment of the present invention. Specific embodiments

[0040] In order to enable those skilled in the art to better understand the solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0041] Refer to Figure 1 , Figure 1 It is a schematic flow chart of a detection method for the circumferential surface of a wheel of a rail vehicle provided by an embodiment of the present application; the detection method for the circumferential surface of the wheel may include:

[0042] S101: Collect the current fluctuation information of the undisturbed circumferential surface undulation of the wheel.

[0043] The fluctuation information of the wheel in this embodiment mainly includes the harmonic number (i.e., order) and wave depth of the circumferential surface of the wheel, which is mainly the undulating surface waveform formed by the unevenness of the circumferential surface of the wheel.

[0044] Refer to Figure 2 and Figure 3 , Figure 2 It is a schematic structural diagram of a measuring device for measuring the fluctuation information of a wheel provided by this embodiment; Figure 3 is Figure 2 A cross-sectional schematic diagram of the contact part between the measuring device and the wheel in

[0045] In Figure 2 and Figure 3 The measuring device shown includes a clamping sleeve 2 that can be clamped to the undisturbed circumferential surface of the wheel 1, and a laser rangefinder 3 provided on the clamping sleeve 2. By measuring the distance information between each point on the circumferential surface of the wheel 1 and the laser rangefinder 3, the fluctuation information within the section of the wheel 1 measured by the measuring device can be obtained.

[0046] In Figure 2In the illustrated embodiment, the measuring device can measure the waveform information of 1 / 4 of the circumferential surface of the wheel 1 at one time. During the measurement process, the measuring device can be used to collect only the waveform information of the unobstructed 1 / 4 of the circumference of the wheel 1 as the current waveform information, or the measuring device can be slid on the unobstructed circumferential surface of the wheel 1 to measure the fluctuation information of all the unobstructed circumferential surfaces of the wheel 1 as the current fluctuation information, so as to improve the accuracy of determining the fluctuation information of the entire wheel 1 subsequently. Of course, the clamping sleeve 2 of the measuring device is not necessarily equal to the arc length of 1 / 4 of the circumference of the wheel 1, and it can also be smaller, which is not limited in this application.

[0047] S102: According to the current fluctuation information and the pre-created correspondence model between the fluctuation information and the vibration data of the wheel, determine the theoretical vibration data of the wheel.

[0048] When there are irregularities on the circumferential surface of the wheel, during the operation of the wheel, the wheel will inevitably vibrate, and the vibration frequency, vibration amplitude, etc. of the wheel are directly affected by the waveform of the undulation of the circumferential surface of the wheel. It can be seen that there should be a certain correlation between the waveform information and the vibration data of the wheel. Therefore, in this embodiment, a correspondence model representing the correlation between the fluctuation information and the vibration data of the wheel is created in sequence.

[0049] This correspondence model can be determined based on feature analysis of the waveform information and the vibration data, or can be determined by performing mathematical operations on the waveform information and the vibration data, which is not limited in this embodiment.

[0050] Optionally, the process of pre-creating this correspondence model may include:

[0051] Collect vibration data samples during the operation of the wheel, and fluctuation information samples of the entire circumferential surface of the wheel;

[0052] Perform neural network training based on the vibration data samples and the fluctuation information samples respectively corresponding to each position point on the wheel when it rotates to fit the track, and obtain the correspondence model.

[0053] The vibration data samples can be the vibration data collected and recorded in real time by vibration sensors installed on the wheel during the actual operation of the vehicle. However, considering that the running speed of the wheel, the bending change of the track, and the height and low fluctuation change of the track may all affect the measured vibration data to a certain extent. Therefore, when selecting the vibration data samples, vibration data samples collected during the operation of the wheel on the same section of the running road at approximately the same speed can be specifically selected. For example, the vibration data during the slow driving of the vehicle when entering the station can be specifically selected as the vibration data samples; in short, factors other than the waveform information should be avoided as much as possible from affecting the vibration of the wheel.

[0054] S103: Find a set of historical vibration data segments in the historical vibration data that has the highest similarity to the theoretical vibration data, and use the vibration data segments adjacent to the historical vibration data segments in the historical vibration data as the predicted vibration data of the wheel.

[0055] Among them, the historical vibration data can be the vibration data collected during the operation of other wheels of the same model as the currently detected wheel.

[0056] To ensure the accuracy of the predicted vibration data determined by searching in the historical vibration data, in another optional embodiment, it can further include:

[0057] Compare the historical fluctuation information corresponding to the historical vibration data with the current fluctuation information of the wheel, and search to obtain multiple segments of historical fluctuation information segments whose similarity to the current fluctuation information is not lower than the first set similarity;

[0058] Compare the similarity of the historical vibration data segments in the historical vibration data corresponding to each segment of historical fluctuation information segment with the theoretical vibration data, and determine a set of historical vibration data segments with the highest similarity to the theoretical vibration data.

[0059] It should be noted that when determining the similarity between the current fluctuation information and the historical fluctuation information, the similarity calculation can be performed based on various different aspects of information such as the harmonic number, wave depth, and waveform change law corresponding to adjacent position points of the current fluctuation information and the historical fluctuation information. The weighted sum can be calculated according to the ratio of the difference between the harmonic numbers of the current fluctuation information and the historical fluctuation information to the harmonic number of the current fluctuation information, and the ratio of the wave depth difference to the wave depth in the current fluctuation information. The smaller the summation result, the higher the similarity, and vice versa, the larger the summation result, the lower the similarity.

[0060] For the similarity between the historical vibration data and the theoretical vibration data, the similarity judgment can be jointly carried out in terms of vibration amplitude, vibration period, and the change law of vibration amplitude, etc. For the determination method of similarity, it can refer to the currently conventional similarity determination method between a set of multiple parameters and another set of multiple parameters. Regarding this, it is not elaborated in detail in this application.

[0061] Generally, for wheels of the same model, as the running time extends, within the entire service life of the wheel, the change in the fluctuation information on the circumferential surface of the wheel is also approximate. Thus, when a historical vibration data segment that is relatively similar to the theoretical vibration data corresponding to the current fluctuation information is found in the historical vibration data, this historical vibration data segment can be regarded as the vibration data corresponding to the current fluctuation information, and the historical vibration data of the adjacent segment of this historical vibration data segment can be regarded as the vibration data corresponding to the circumferential region of the wheel where the unmeasured waveform information corresponding to the current fluctuation information is located. Thus, it is equivalent to obtaining the vibration data of a complete circumferential surface of the wheel.

[0062] Optionally, in another optional embodiment of the present application, the process of determining the estimated vibration data corresponding to the occluded circumferential surface on the wheel may include:

[0063] After determining the historical vibration data segment in the historical vibration data that is closest to the theoretical vibration data corresponding to the current fluctuation data,

[0064] Taking the vibration data segment with a set segment length adjacent to the historical vibration data segment in the historical vibration data as the estimated vibration data; wherein, the set segment length is determined according to the wheel speed corresponding to the collection of historical vibration data.

[0065] It can be understood that the sum of the segment length corresponding to the historical vibration data segment corresponding to the current fluctuation information and the above set segment length should exactly equal the segment of the historical vibration data corresponding to one full rotation of the wheel. Obviously, this set segment length is related to the wheel occlusion rate.

[0066] S104: Determine the estimated fluctuation information corresponding to the circumferential surface of the wheel except for the circumferential surface corresponding to the current fluctuation information based on the estimated vibration data, and determine the fluctuation information of the complete circumferential surface of the wheel based on the estimated fluctuation information and the current fluctuation information.

[0067] As described above, the estimated vibration data can be regarded as the vibration data corresponding to the circumferential surface of the wheel where the waveform information is occluded and not measured. Therefore, substituting this estimated vibration data into the corresponding relationship model can determine the unmeasured waveform information of the wheel. Obviously, the measured waveform information and the current waveform information can be spliced with each other to obtain the waveform information of a complete wheel.

[0068] Of course, this estimated vibration data is data within a segment of the historical vibration data. If there is corresponding historical waveform information in this historical vibration data, obviously, the historical waveform information corresponding to this estimated vibration data can be directly used as the waveform information of the unmeasured circumferential region of the wheel, and splicing this historical waveform information with the current wheel information can also obtain the fluctuation information of the complete circumferential surface of the wheel.

[0069] In summary, in the present application, by utilizing the characteristic that there is a corresponding correlation between the undulation of the circumferential surface of the wheel and the vibration data during the operation of the wheel, in the braking state where the wheel is directly in contact with the rail, only the current undulation information of a part of the circumferential surface of the wheel is collected. Then, through the correlation relationship model between the undulation information and the vibration data, the theoretical vibration data corresponding to the current undulation information can be determined. By using the theoretical vibration data that is closest to the estimated vibration data corresponding to the circumferential surface of the wheel in the historical vibration data, based on the estimated vibration data, the waveform information of the circumferential surface of the wheel that cannot be measured can be predicted. Then, by combining the currently measured current undulation information, the undulation information of the entire circumferential surface of the wheel can be obtained. During the entire detection process, no additional tools such as jacks are required, reducing the consumption of manpower and material resources and lowering the detection difficulty of the circumferential surface information of the wheel.

[0070] Next, the detection device for the circumferential surface of the wheel of the rail vehicle provided in the embodiments of the present invention will be introduced. The detection device for the circumferential surface of the wheel of the rail vehicle described below can be mutually referred to corresponding to the detection method for the circumferential surface of the wheel of the rail vehicle described above.

[0071] Figure 4 is the structural block diagram of the detection device for the circumferential surface of the wheel of the rail vehicle provided in the embodiments of the present invention. Refer to Figure 4 The detection device for the circumferential surface of the wheel of the rail vehicle in

[0072] The data acquisition module 100 is used to acquire the current undulation information of the unobstructed circumferential surface undulation of the wheel;

[0073] The first operation module 200 is used to determine the theoretical vibration data of the wheel according to the current undulation information and the pre-created correspondence relationship model between the undulation information and the vibration data of the wheel;

[0074] The second operation module 300 is used to find a set of historical vibration data segments with the highest similarity to the theoretical vibration data in the historical vibration data, and use the vibration data segment adjacent to the historical vibration data segment in the historical vibration data as the estimated vibration data of the wheel;

[0075] The third operation module 400 is used to determine the estimated undulation information corresponding to the circumferential surface of the wheel except for the circumferential surface corresponding to the current undulation information according to the estimated vibration data, and determine the undulation information of the entire circumferential surface of the wheel according to the estimated undulation information and the current undulation information.

[0076] In an optional embodiment of the present application, it further includes a model creation module, which is used to collect vibration data samples during the operation of the wheel and fluctuation information samples on the entire circumferential surface of the wheel; and perform neural network training based on the vibration data samples and fluctuation information samples respectively corresponding to each position point on the wheel when it rotates to fit the track, so as to obtain the corresponding relationship model.

[0077] In an optional embodiment of the present application, the second operation module 300 is specifically configured to use, as the predicted vibration data, a vibration data segment with a set segment length adjacent to the historical vibration data segment in the historical vibration data; wherein, the set segment length is determined according to the wheel speed corresponding to the collection of the historical vibration data.

[0078] In an optional embodiment of the present application, the third operation module 400 is used to obtain the predicted fluctuation information on the circumferential surface of the wheel blocked according to the predicted vibration data and the corresponding relationship model.

[0079] In an optional embodiment of the present application, the second operation module 300 is specifically configured to compare the historical fluctuation information corresponding to the historical vibration data with the current fluctuation information of the wheel, and search for multiple historical fluctuation information segments whose similarity to the current fluctuation information is not lower than the first set similarity; compare the similarity between the historical vibration data segments in the historical vibration data corresponding to each historical fluctuation information segment and the theoretical vibration data, and determine a group of historical vibration data segments with the highest similarity to the theoretical vibration data.

[0080] The detection device for the circumferential surface of the wheel of the rail vehicle in this embodiment is used to implement the foregoing detection method for the circumferential surface of the wheel of the rail vehicle. Therefore, the specific implementation manners in the detection device for the circumferential surface of the wheel of the rail vehicle can be seen in the embodiment part of the detection method for the circumferential surface of the wheel of the rail vehicle in the foregoing text. For example, the data acquisition module 100, the first operation module 200, the second operation module 300, and the third operation module 400 are respectively used to implement steps S101, S102, S103, and S104 in the foregoing detection method for the circumferential surface of the wheel of the rail vehicle. Therefore, the specific implementation manners can be referred to the descriptions of the corresponding various part embodiments and will not be elaborated herein.

[0081] As Figure 5 shown, Figure 5 It is a structural block diagram of the detection device for the circumferential surface of the wheel of the rail vehicle provided by the embodiment of the present invention. In an embodiment of the detection device for the circumferential surface of the wheel of the rail vehicle provided by the present application, the detection device for the circumferential surface of the wheel of the rail vehicle may include:

[0082] A memory 10 for storing a computer program;

[0083] A processor 20 for executing the computer program to implement the steps of the method for detecting the circumferential surface of the wheel of the rail vehicle described in any one of the preceding items.

[0084] The steps of the method for detecting the circumferential surface of the wheel of the rail vehicle executed by the processor may include:

[0085] Collecting the current fluctuation information of the undisturbed circumferential surface undulation of the wheel;

[0086] According to the current fluctuation information and the correspondence model between the pre-created fluctuation information and the vibration data of the wheel, determining the theoretical vibration data of the wheel;

[0087] Searching in the historical vibration data for a set of historical vibration data segments with the highest similarity to the theoretical vibration data, and using the vibration data segments adjacent to the historical vibration data segments in the historical vibration data as the estimated vibration data of the wheel;

[0088] Determining the estimated fluctuation information corresponding to the circumferential surface of the wheel except for the circumferential surface corresponding to the current fluctuation information according to the estimated vibration data, and determining the fluctuation information of the complete circumferential surface of the wheel according to the estimated fluctuation information and the current fluctuation information.

[0089] This application also provides a computer-readable storage medium, in which a computer program is stored, and the computer program is executed to implement the steps of the method for detecting the circumferential surface of the wheel of the rail vehicle described in any one of the preceding items.

[0090] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other. For the devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple. For the relevant parts, refer to the description in the method part.

[0091] Those skilled in the art can further realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the components and steps of the examples have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

Claims

1. A method for detecting the circumferential surface of a wheel of a railway vehicle, characterized in that: include: Collecting current fluctuation information of the undulation of the unobstructed circumferential surface of the wheel; Determining the theoretical wheel vibration data based on the current fluctuation information and a pre-created correspondence model between the fluctuation information and the wheel vibration data; searching, in the historical vibration data, for a group of historical vibration data segments having the highest similarity to the theoretical vibration data, and using, in the historical vibration data, vibration data segments adjacent to the historical vibration data segments as the estimated vibration data of the wheel; determining estimated fluctuation information corresponding to a circumferential surface of the wheel other than the circumferential surface corresponding to the current fluctuation information based on the estimated vibration data, and determining fluctuation information of the complete circumferential surface of the wheel based on the estimated fluctuation information and the current fluctuation information; The process of pre-creating the corresponding relationship model includes: Collecting vibration data samples during the operation of the wheel and fluctuation information samples of the entire circumferential surface of the wheel; A neural network training is performed based on the vibration data samples and fluctuation information samples corresponding to each position point on the wheel when the wheel rotates to fit the track to obtain the corresponding relationship model.

2. The method for detecting the circumferential surface of a wheel of a rail vehicle according to claim 1, wherein: Using a vibration data segment adjacent to the historical vibration data segment in the historical vibration data as the estimated vibration data of the wheel, including: A vibration data segment of a set segment length adjacent to the historical vibration data segment in the historical vibration data is used as the estimated vibration data; wherein the set segment length is determined according to the wheel speed corresponding to when the historical vibration data is collected.

3. The method for detecting the circumferential surface of a wheel of a railway vehicle according to claim 1, wherein: Determining, based on the estimated vibration data, estimated fluctuation information corresponding to a circumferential surface of the wheel other than the circumferential surface corresponding to the current fluctuation information, includes: According to the estimated vibration data and the corresponding relationship model, estimated fluctuation information of the obscured circumferential surface of the wheel is obtained.

4. The method for detecting the circumferential surface of a wheel of a railway vehicle according to any one of claims 1 to 3, characterized in that: Searching the historical vibration data for a group of historical vibration data segments having the highest similarity to the theoretical vibration data, including: Comparing the historical fluctuation information corresponding to the historical vibration data with the current fluctuation information of the wheel, and searching for a plurality of historical fluctuation information segments having a similarity with the current fluctuation information that is not less than a first set similarity; The historical vibration data segments in the historical vibration data corresponding to each of the historical fluctuation information segments are compared with the theoretical vibration data for similarity, and a group of the historical vibration data segments having the highest similarity with the theoretical vibration data is determined.

5. A device for detecting the circumferential surface of a wheel of a railway vehicle, characterized in that: include: A data acquisition module, used to collect current fluctuation information of the undulations of the unobstructed circumferential surface of the wheel; a first operation module, configured to determine the theoretical wheel vibration data based on the current fluctuation information and a pre-created correspondence relationship model between the fluctuation information and the wheel vibration data; a second operation module, configured to search the historical vibration data for a group of historical vibration data segments having the highest similarity to the theoretical vibration data, and use vibration data segments adjacent to the historical vibration data segments in the historical vibration data as the estimated vibration data of the wheel; a third computing module, configured to determine, based on the estimated vibration data, estimated fluctuation information corresponding to a circumferential surface of the wheel other than the circumferential surface corresponding to the current fluctuation information, and determine fluctuation information of the complete circumferential surface of the wheel based on the estimated fluctuation information and the current fluctuation information; It also includes a model creation module for collecting vibration data samples during the operation of the wheel and fluctuation information samples of the entire circumferential surface of the wheel; neural network training is performed based on the vibration data samples and fluctuation information samples corresponding to each position point on the wheel when it rotates to fit the track to obtain the corresponding relationship model.

6. The device for detecting the circumferential surface of a wheel of a railway vehicle according to claim 5, wherein: The second operation module is specifically used to compare the historical fluctuation information corresponding to the historical vibration data with the current fluctuation information of the wheel, and to search for multiple historical fluctuation information segments whose similarity with the current fluctuation information is not less than a first set similarity; and to compare the historical vibration data segments in the historical vibration data corresponding to each segment of the historical fluctuation information with the theoretical vibration data for similarity, and to determine a group of historical vibration data segments with the highest similarity with the theoretical vibration data.

7. A device for detecting the circumferential surface of a wheel of a railway vehicle, characterized in that: include: memory for storing computer programs; A processor is used to execute the computer program to implement the steps of the method for detecting the circumferential surface of the wheel of a railway vehicle as described in any one of claims 1 to 4.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which is executed to implement the steps of the method for detecting the circumferential surface of a wheel of a rail vehicle according to any one of claims 1 to 4.

Images