Steel rail corrugation degree grading method, device, equipment, medium and program

Abstract

The invention relates to a steel rail corrugation degree grading method, device, equipment, medium and program, and the method comprises the steps: collecting the vertical vibration information of a train axle box, and constructing a vertical vibration data set according to the vertical vibration information; performing low-pass filtering on the vertical vibration data set to obtain a filtered data set, and calculating a track impact index corresponding to the filtered data set; judging whether a rail corrugation section exists or not according to the rail impact index and a preset over-limit rail impact index threshold value; if the steel rail corrugation section exists, corrugation information corresponding to the steel rail corrugation section is used for generating a corrugation data set; and extracting signal time domain features corresponding to the corrugation data set, and performing full-connection mapping on the signal time domain features to obtain a disease degree grading result.

Description

Technical Field

[0001] The present disclosure relates to the field of deep learning technology, and in particular to a rail corrugation disease degree grading method, device, equipment, medium and program. Background Art

[0002] Rail corrugation refers to a periodic, wave-shaped, uneven wear along the longitudinal surface of the rail. When a vehicle passes through the rail corrugation section, it will generate large vibrations, causing abnormal wheel-rail forces, resulting in fatigue damage to various structural components such as vehicles, tracks, and fasteners. In severe cases, it will endanger the safety of train operation.

[0003] There are many rail corrugation detection methods in the prior art, which can be divided into two categories: ground detection methods and vehicle-mounted detection methods. For example, the ground detection method often uses static measurement methods such as corrugation detectors and steel rulers. Although the detection accuracy is high and accurate rail surface roughness data can be directly obtained, the detection efficiency is low and the detection distance is limited. It is impossible to cover long-distance tracks of heavy-duty railways all day long; vehicle-mounted detection only points out how to detect corrugation damage, but does not clearly point out the degree of corrugation damage. In addition, most of the existing vehicle-mounted detections are collected on high-speed railways and subways. It is difficult to obtain the original data of heavy-duty railway trains, resulting in inaccurate in-vehicle vibration data detection results. In summary, the prior art has problems such as low detection efficiency, inaccurate detection data, and low grading of the degree of corrugation damage. Summary of the invention

[0004] The present invention provides a rail corrugation disease degree classification method, device, equipment, medium and program, the main purpose of which is to solve the problems of low efficiency and inaccurate detection data of heavy-load railway rail corrugation.

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

[0006] Collecting vertical vibration information at a train axle box, and constructing a vertical vibration data set according to the vertical vibration information;

[0007] Performing low-pass filtering on the vertical vibration data set to obtain a filtered data set, and calculating a track impact index corresponding to the filtered data set;

[0008] Determining whether there is a rail corrugation section according to the track impact index and a preset over-limit track impact index threshold;

[0009] If there is a rail corrugation section, generating a corrugation data set with corrugation information corresponding to the rail corrugation section;

[0010] The signal time domain features corresponding to the corrugation data set are extracted, and the signal time domain features are fully connected and mapped to obtain a disease degree classification result.

[0011] In some embodiments, constructing a vertical vibration dataset according to the vertical vibration information includes:

[0012] Performing data cleaning and noise reduction processing on the vertical vibration information to obtain pollution-free vertical vibration information;

[0013] Generating information retrieval of each vertical vibration information in a preset database according to the pollution-free vertical vibration information;

[0014] The vertical vibration information is associated and stored according to the information retrieval to obtain a vertical vibration data set.

[0015] In some embodiments, performing low-pass filtering on the vertical vibration data set to obtain a filtered data set includes:

[0016] calculating a stopband ripple and a low-pass filter order based on the vertical vibration data set;

[0017] A low-pass filter is designed according to the stop-band ripple and the order of the low-pass filter, and the vertical vibration data set is low-pass filtered by the low-pass filter to obtain a filtered data set.

[0018] In some embodiments, judging whether there is a rail corrugation section according to the rail impact index and a preset overrun rail impact index threshold value includes:

[0019] If the rail impact index is not within the rail impact index threshold range, it is determined that there is no rail corrugation section in the measured section;

[0020] If the track impact index is within the track impact index threshold range, it is determined that there is a rail corrugation section in the measured section.

[0021] In some embodiments, combining the corrugation information corresponding to the rail corrugation section into a corrugation data set includes:

[0022] dividing the vertical vibration data set into a plurality of sliding windows according to a preset window length;

[0023] A sliding window containing corrugation damage is identified by a preset rail impact index method, and a corrugation data set is constructed according to information in the sliding window containing corrugation damage.

[0024] In some embodiments, extracting the signal time domain features corresponding to the corrugation data set includes:

[0025] Performing convolution, pooling and activation processing on the corrugation data set to obtain a processed corrugation data set;

[0026] The processed corrugation data set is input into a pre-acquired long short-term memory network to obtain signal time domain features.

[0027] In some embodiments, inputting the processed corrugation data set into a long short-term memory network to obtain signal time domain features includes:

[0028] The signal sequences of the forget gate, input gate and output gate are calculated according to the preset signal input sequence and bias parameters;

[0029] The time domain features of the signal input sequence are extracted according to each gate signal sequence and a preset weight matrix.

[0030] In a second aspect, the present disclosure provides a rail corrugation disease degree grading device, comprising:

[0031] An information collection module collects vertical vibration information at the axle box of the train and constructs a vertical vibration data set according to the vertical vibration information;

[0032] A track impact index calculation module performs low-pass filtering on the vertical vibration data set to obtain a filtered data set, and calculates a track impact index corresponding to the filtered data set;

[0033] A judgment module, for judging whether there is a rail corrugation section according to the track impact index and a preset over-limit track impact index threshold;

[0034] A corrugation data set generating module, if there is a rail corrugation section, generates a corrugation data set with the corrugation information corresponding to the rail corrugation section;

[0035] The corrugation damage degree grading module extracts the signal time domain features corresponding to the corrugation data set, performs full connection mapping on the signal time domain features, and obtains the damage degree grading result.

[0036] In a third aspect, the present disclosure provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method described in the above aspects.

[0037] In a fourth 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.

[0038] In a fifth 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.

[0039] In the embodiment of the present invention, by collecting the vertical vibration information of the train axle box to construct a vertical vibration data set, the high quality of the data set can be ensured to facilitate the accuracy of subsequent processing and analysis; the vertical vibration data set is low-pass filtered, and the corresponding track impact index after low-pass filtering is calculated, which is helpful to timely discover and deal with rail corrugation problems, improve the detection efficiency of signal processing, and ensure the safety of the track to the greatest extent; the corrugation information corresponding to the rail corrugation section is combined into a corrugation data set, the corresponding signal time domain features are extracted, and the signal time domain features are fully connected and mapped, which can reduce the feature dimension and thus reduce the amount of calculation, thereby obtaining a disease degree classification result. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The present disclosure will be described in more detail below based on embodiments and with reference to the accompanying drawings:

[0041] Figure 1 A schematic flow chart of a method for grading the degree of rail corrugation damage provided in an embodiment of the present disclosure.

[0042] Figure 2 A schematic structural diagram of a rail corrugation disease degree grading method and device provided in an embodiment of the present disclosure.

[0043] Figure 3 An embodiment of the present disclosure provides a vibration signal of an axle box including corrugation.

[0044] Figure 4 It is the track impact index calculated by using the axle box vibration signal in the embodiment of the present disclosure.

[0045] Figure 5 This is a diagram of the LSTM network structure provided for an embodiment of the present disclosure.

[0046] In the drawings, the same reference numerals are used for the same components, and the drawings are not drawn to scale. DETAILED DESCRIPTION

[0047] 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.

[0048] 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.

[0049] An embodiment of the present application provides a method for grading the degree of rail corrugation. The execution subject of the method for grading the degree of rail corrugation includes but is not limited to a server, a terminal, and at least one of the electronic devices that can be configured to execute the system provided in the embodiment of the present application. In other words, the method for grading the degree of rail corrugation can be executed by software or hardware installed on a terminal device or a server device. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0050] Reference Figure 1 , which is a schematic flow chart of a rail corrugation disease degree classification method provided by an embodiment of the present invention.

[0051] In this embodiment, the rail corrugation damage degree classification method includes:

[0052] S1. Collect vertical vibration information at a train axle box, and construct a vertical vibration data set according to the vertical vibration information.

[0053] In the embodiment of the present invention, the vertical vibration information refers to the vibration of the axle box, i.e., the component connecting the wheels and axles, in the vertical direction during the operation of the train, including the vertical vibration acceleration of the axle box, the depth of the corrugated valley, and the vibration characteristics of the axle box bearing of the rail vehicle under the coupling of internal and external excitations.

[0054] In the embodiment of the present invention, the vertical vibration information can be obtained by a vibration acceleration sensor deployed at the axle box of the train. During the acquisition process, the electrical signal acquired by the sensor exists in a continuous form, and the electrical signal is converted into a digital signal form, i.e., the axle box vibration acceleration signal for acquisition. In the embodiment of the present invention, in order to apply the vertical vibration information more quickly and intuitively, a data set needs to be constructed.

[0055] In an embodiment of the present invention, constructing a vertical vibration data set according to the vertical vibration information includes:

[0056] Performing data cleaning and noise reduction processing on the vertical vibration information to obtain pollution-free vertical vibration information;

[0057] Generating information retrieval of each vertical vibration information in a preset database according to the pollution-free vertical vibration information;

[0058] The vertical vibration information is associated and stored according to the information retrieval to obtain a vertical vibration data set.

[0059] In detail, the data cleaning refers to identifying and processing abnormal information, removing the abnormal information and correcting it; the noise reduction processing refers to denoising the information containing noise pollution to eliminate noise interference to the greatest extent.

[0060] In the embodiment of the present invention, abnormal information is identified and processed through data cleaning to avoid deviations in subsequent analysis; the present invention reduces noise pollution to a minimum through noise reduction processing, thereby improving the accuracy of the data. The pollution-free vertical vibration information is used to generate information retrieval of each vertical vibration information in a preset database and associated storage is performed to obtain a vertical vibration data set, which is convenient for subsequent data search and analysis and improves data processing efficiency.

[0061] In the embodiment of the present invention, the vertical vibration information at the axle box of the train is collected to ensure the integrity of the vertical vibration information, so as to understand the stability of the train operation and the quality of the track, so as to facilitate the accuracy of subsequent processing and analysis.

[0062] S2. Low-pass filtering is performed on the vertical vibration data set to obtain a filtered data set, and a track impact index corresponding to the filtered data set is calculated.

[0063] In the embodiment of the present invention, the low-pass filtering is a filtering method that allows signals with a frequency lower than a set cutoff frequency to pass normally, while signals with a frequency higher than the set cutoff frequency are blocked and weakened.

[0064] The present invention can design a low-pass filter by calculating the stopband cutoff frequency, stopband ripple, low-pass filter order, stopband maximum attenuation and other parameters of the vertical vibration data set, and then calculate the low-pass filter according to the low-pass filter. The low-pass filter only allows signals below a certain frequency to pass through the filter without attenuation.

[0065] Specifically, the vertical vibration data set is low-pass filtered to obtain a filtered data set, including:

[0066] calculating a stopband ripple and a low-pass filter order based on the vertical vibration data set;

[0067] A low-pass filter is designed according to the stop-band ripple and the order of the low-pass filter, and the vertical vibration data set is low-pass filtered by the low-pass filter to obtain a filtered data set.

[0068] Specifically, the stopband ripple can be calculated using the following formula:

[0069]

[0070] Among them, A s It represents the preset stopband cutoff frequency, and ∈ represents the stopband ripple.

[0071] Specifically, the low-pass filter order can be calculated using the following formula, where the low-pass filter order is rounded up:

[0072]

[0073] Where N represents the order of the low-pass filter, A s Indicates the preset stopband cutoff frequency, A p It represents the preset stopband maximum attenuation, and ∈ represents the stopband ripple.

[0074] The present invention designs a low-pass filter by calculating the parameters, performs low-pass filtering using the low-pass filter, removes short-term fluctuations in the signal, thereby retaining the long-term development trend and providing a smooth form of the signal. In the actual signal processing process, noise can be removed, the quality of the signal can be improved, and the stability and reliability of the signal can be ensured.

[0075] Specifically, the following formula can be used for low-pass filtering to obtain a filtered data set:

[0076]

[0077] In detail, y(n) is the filtered data set, {x(n)|n=1,2…} is the original vertical vibration data set, n is the total number of original vertical vibration data, a(k) and b(j) represent low-pass filter parameters, and capital letters N and P represent the low-pass filter order.

[0078] In the embodiment of the present invention, the low-pass filter can filter out high-frequency interference in sensor signal processing, improve the accuracy of sensor signals, thereby improving the efficiency and effect of signal processing, and facilitate the subsequent judgment of the presence of corrugation and calculation of the track impact index.

[0079] The present invention uses the following formula to calculate the orbital impact index corresponding to the filtered data set:

[0080]

[0081] Among them, y(n) is the filtered data set, that is, the axle box acceleration signal after low-pass filtering, K represents the preset moving window length, t TⅡ is the orbital impact index, S r is the effective value of the axle box acceleration after low-pass filtering at the measuring point r, For S r The average value of .

[0082] In the embodiment of the present invention, the calculation of the rail impact index can be used as a criterion for determining whether rail corrugation exists, providing important data support for subsequent rail maintenance and operation. By comparing the rail impact index before and after maintenance, the maintenance effect can be evaluated, and the occurrence and development of rail corrugation diseases can be effectively understood.

[0083] In the embodiment of the present invention, the vertical vibration data set is low-pass filtered to obtain a filtered data set, and the track impact index corresponding to the filtered data set is calculated to grasp the maximum extent of rail corrugation damage, which helps to timely discover and deal with rail corrugation problems, improves the efficiency of signal processing, and ensures the safety of the track to the greatest extent.

[0084] S3. Determine whether there is a rail corrugation section according to the track impact index and a preset over-limit track impact index threshold.

[0085] In the embodiment of the present invention, judging whether there is a rail corrugation section according to the track impact index and a preset overrun track impact index threshold value includes:

[0086] If the rail impact index is not within the rail impact index threshold range, it is determined that there is no rail corrugation section in the measured section;

[0087] If the track impact index is within the track impact index threshold range, it is determined that there is a rail corrugation section in the measured section.

[0088] In detail, shown Figure 3 This is a vibration signal of the axle box including corrugation. Figure 4 is the rail impact index calculated using the axle box vibration signal.

[0089] In the embodiment of the present invention, by measuring and analyzing the vibration signal of the axle box containing corrugation, it is quickly and accurately determined whether the measured section has corrugation damage, thereby determining the location of the corrugation, reducing the subsequent calculation amount, improving maintenance efficiency, and reducing maintenance costs.

[0090] S4. If there is a rail corrugation section, the corrugation information corresponding to the rail corrugation section is combined into a corrugation data set.

[0091] In an embodiment of the present invention, combining the corrugation information corresponding to the rail corrugation section into a corrugation data set includes:

[0092] dividing the vertical vibration data set into a plurality of sliding windows according to a preset window length;

[0093] A sliding window containing corrugation damage is identified by a preset rail impact index method, and a corrugation data set is constructed according to information in the sliding window containing corrugation damage.

[0094] In detail, the track impact index method is a method for judging short-wave irregularity of railway tracks, which mainly focuses on evaluating the severity of the disease, including train speed, wheel-rail clearance, wheelset wear, etc. The track impact index method can better maintain and manage the track, realize the effective control and treatment of short-wave irregularity of railway tracks, and thus improve the accuracy and reliability of information.

[0095] In detail, the original vertical vibration data set is {x(n)|n=1,2…}. According to the preset window length K, the original data set can be divided into several sliding windows. The data in each sliding window is {x(iK)~x((i+1)K)|i=0,1…}. If the existence of corrugation in the jth sliding window is detected according to the orbital impact index method, a new data set {x(jK)~x((j+1)K)|j=0,1…} is constructed, where j is the sliding window index of the identified corrugation.

[0096] In an embodiment of the present invention, if there is a rail corrugation section, the corrugation information corresponding to the rail corrugation section is combined into a corrugation data set, and the corrugation data set includes all vibration data containing corrugation diseases in the original data set, thereby reducing the amount of calculation for subsequent network training and improving the efficiency of corrugation detection.

[0097] S5. Extracting the signal time domain features corresponding to the corrugation data set, performing full connection mapping on the signal time domain features, and obtaining a disease degree classification result.

[0098] In the embodiment of the present invention, the signal time domain features are mainly features of the signal on the time axis, which refers to convolutional neural network learning of the corrugation data set, including features such as amplitude, frequency, and phase of the signal.

[0099] In an embodiment of the present invention, extracting the signal time domain features corresponding to the corrugation data set includes:

[0100] Performing convolution, pooling and activation processing on the corrugation data set to obtain a processed corrugation data set;

[0101] The processed corrugation data set is input into a pre-acquired long short-term memory network to obtain signal time domain features.

[0102] In detail, the convolutional neural network includes a convolution module and a fully connected module, the convolution module extracts features, and the fully connected module classifies the features; the present invention is mainly used to reduce feature dimensions and reduce feature dimensionality by performing convolution, activation function and pooling processing on the wave grinding data set.

[0103] In the embodiment of the present invention, the processed corrugation data set is input into a long short-term memory network, i.e., an LSTM network, to obtain the signal time domain features. The LSTM network is a time cycle network. Since the vibration response caused by rail corrugation is periodic, the LSTM network is more suitable for extracting signal time domain features. The LSTM network structure diagram is shown in FIG. Figure 5 shown.

[0104] In the implementation of the present invention, the processed corrugation data set is input into a long short-term memory network to obtain the signal time domain features, including:

[0105] The signal sequences of the forget gate, input gate and output gate are calculated according to the preset signal input sequence and bias parameters;

[0106] The time domain features of the signal input sequence are extracted according to each gate signal sequence and a preset weight matrix.

[0107] In the embodiment of the present invention, the core of the LSTM network is a signal sequence. The time domain features of the signal sequence are extracted through three gate structures: a forget gate, an input gate, and an output gate. The expression is shown in the formula:

[0108] f t =σ(W f *[h t-1 ,x t ]+b f )

[0109] i t =σ(W i *[h t-1 ,x t ]+b i )

[0110]

[0111] o t=σ(W o *[h t-1 ,x t ]+b o )

[0112] h t =o t *tanh(C t )

[0113] Among them C t , C t-1 Represent the current and previous signal sequences respectively, h t 、h t-1 Represent the current output sequence and the previous output sequence respectively, x t represents the input sequence, f t represents the output sequence, i t Indicates the update degree of the input gate, o t represents a neural unit, Represents the data signal of the control input, W f , W i , W c , W o are the weight matrices of the forget gate, input modulation gate, input gate, and output gate respectively; b f , b i , b c , b o They are the bias parameters of the forget gate, input modulation gate, input gate, and output gate respectively.

[0114] In an embodiment of the present invention, the local features of the corrugated grinding data set are captured by a convolution layer, and the local features are transformed nonlinearly using an activation function to enhance the expression ability of the signal sequence. The local features are then reduced in dimension using a pooling layer to reduce the amount of calculation and extract key features, thereby laying a foundation for subsequent further analysis.

[0115] In an embodiment of the present invention, in order to more accurately capture the temporal dependency of the signal and overcome the gradient vanishing problem encountered by the convolutional neural network when processing long time series, the preprocessed corrugated grinding data set is input into the LSTM network to extract the time domain features of the signal, and then the time domain features are fully connected and mapped, and weighted summation is performed using preset weight coefficients to obtain the disease degree grading result.

[0116] In the embodiment of the present invention, by collecting the vertical vibration information of the train axle box to construct a vertical vibration data set, the high quality of the data set can be ensured to facilitate the accuracy of subsequent processing and analysis; the vertical vibration data set is low-pass filtered, and the corresponding track impact index after low-pass filtering is calculated, which is helpful to timely discover and deal with rail corrugation problems, improve the detection efficiency of signal processing, and ensure the safety of the track to the greatest extent; the corrugation information corresponding to the rail corrugation section is combined into a corrugation data set, the corresponding signal time domain features are extracted, and the signal time domain features are fully connected and mapped, which can reduce the feature dimension and thus reduce the amount of calculation, thereby obtaining a disease degree classification result.

[0117] On the basis of the above embodiments, this embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method described in the above embodiments.

[0118] In some implementations of this embodiment, a computer-readable storage medium is provided, on which a computer program is stored, characterized in that when the computer program is executed by a processor, the steps of the method described in the above embodiment are implemented.

[0119] 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.

[0120] The processor may include, but is not limited to, one or more processors or microprocessors, etc. 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 methods in the above embodiments.

[0121] The computer-readable storage medium may be implemented by any type of volatile or non-volatile storage device or a combination thereof. 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, CDROM, DVDROM, Blu-ray disc, etc.).

[0122] 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.

[0123] 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.

[0124] The processor may communicate with external devices via an I / O bus via a wired or wireless network.

[0125] 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.

[0126] 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.

[0127] 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.

[0128] 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 disease degree classification method, characterized in that: The method comprises: Collecting vertical vibration information at a train axle box, and constructing a vertical vibration data set according to the vertical vibration information; Performing low-pass filtering on the vertical vibration data set to obtain a filtered data set, and calculating a track impact index corresponding to the filtered data set; Determining whether there is a rail corrugation section according to the track impact index and a preset over-limit track impact index threshold; If there is a rail corrugation section, generating a corrugation data set according to the corrugation information corresponding to the rail corrugation section; The signal time domain features corresponding to the corrugation data set are extracted, and the signal time domain features are fully connected and mapped to obtain a disease degree classification result.

2. The method according to claim 1, characterized in that The constructing a vertical vibration data set according to the vertical vibration information includes: Performing data cleaning and noise reduction processing on the vertical vibration information to obtain pollution-free vertical vibration information; Generating information retrieval of each vertical vibration information in a preset database according to the pollution-free vertical vibration information; The vertical vibration information is associated and stored according to the information retrieval to obtain a vertical vibration data set.

3. The method according to claim 1, characterized in that The step of low-pass filtering the vertical vibration data set to obtain a filtered data set includes: calculating a stopband ripple and a low-pass filter order based on the vertical vibration data set; A low-pass filter is designed according to the stop-band ripple and the order of the low-pass filter, and the vertical vibration data set is low-pass filtered by the low-pass filter to obtain a filtered data set.

4. The method according to claim 1, characterized in that The determining whether there is a rail corrugation section according to the track impact index and a preset overrun track impact index threshold value includes: If the rail impact index is not within the rail impact index threshold range, it is determined that there is no rail corrugation section in the measured section; If the track impact index is within the track impact index threshold range, it is determined that there is a rail corrugation section in the measured section.

5. The method according to claim 1, characterized in that Combining the corrugation information corresponding to the rail corrugation section into a corrugation data set includes: dividing the vertical vibration data set into a plurality of sliding windows according to a preset window length; A sliding window containing corrugation damage is identified by a preset rail impact index method, and a corrugation data set is constructed according to information in the sliding window containing corrugation damage.

6. The method according to claim 1, characterized in that The extracting of the signal time domain features corresponding to the corrugation data set includes: Performing convolution, pooling and activation processing on the corrugation data set to obtain a processed corrugation data set; The processed corrugation data set is input into a pre-acquired long short-term memory network to obtain signal time domain features.

7. A rail corrugation damage grading device, characterized in that: include: An information collection module collects vertical vibration information at the axle box of the train and constructs a vertical vibration data set according to the vertical vibration information; A track impact index calculation module performs low-pass filtering on the vertical vibration data set to obtain a filtered data set, and calculates a track impact index corresponding to the filtered data set; A judgment module, for judging whether there is a rail corrugation section according to the track impact index and a preset over-limit track impact index threshold; A corrugation data set generating module, if there is a rail corrugation section, generates a corrugation data set with the corrugation information corresponding to the rail corrugation section; The corrugation damage degree grading module extracts the signal time domain features corresponding to the corrugation data set, performs full connection mapping on the signal time domain features, and obtains the damage degree grading result.

8. A computer device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 6.

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

Images