A method and system for detecting aging degree of rubber pads of railway fasteners

Abstract

The present invention discloses a method and system for detecting the aging degree of railway fastener rubber pads. The method comprises: simulating the fastener rubber pads according to the fastener rubber pad refinement model, taking into account the nonlinear influence, track unevenness and different fastener rubber pad aging degrees, obtaining the rail vibration acceleration in the time domain, and constructing the original data set of the fastener rubber pad aging according to the rail vibration acceleration; using adaptive frequency bandwidth and weight adjustment empirical wavelet transform to decompose the vibration response signal of the rail to obtain the modal component, classifying the fastener rubber pad aging degree according to the fastener rubber pad aging degree index, and obtaining the fastener rubber pad aging degree classification result corresponding to the modal component; training the convolutional neural network to obtain the fastener rubber pad aging degree detection model; the fastener rubber pad aging degree detection model outputs the fastener rubber pad aging degree corresponding to the rail vibration acceleration data. The efficiency of the convolutional neural network in detecting the aging degree of the fastener rubber pad can be improved.

Description

Technical Field

[0001] The invention belongs to the technical field of railway fastener rubber pad detection, and in particular relates to a railway fastener rubber pad aging degree detection method and system. Background Art

[0002] As the connecting parts for fixing rails and sleepers, the rubber pad status of the fastener directly affects driving safety. Since the fastener system stiffness is composed of the clamping part stiffness and the rubber pad stiffness, the clamping part stiffness is only about 6% of the fastener rubber pad stiffness. Therefore, the fastener rubber pad stiffness can be approximately regarded as the total stiffness of the entire fastener system. The aging of the fastener rubber pad will directly affect the size of the fastener rubber pad stiffness, so the detection of the aging status of the fastener rubber pad is one of the most important links in the entire infrastructure safety detection. However, the means of detecting the status of rail fasteners mainly rely on the inspection of railway workers along the line. This detection method: the detection efficiency is very low and the labor intensity is high.

[0003] At present, most intelligent detection methods for structural damage are based on computer vision. The limitation of this method is that it cannot identify early damage and invisible defects of the structure, such as aging of rubber pads, invisible cracks in fasteners in the fastener system, loose bolts in the fastener system, etc. Therefore, the change in stiffness caused by the aging of fastener rubber pads will cause changes in rail acceleration. Summary of the invention

[0004] The present invention provides a method and system for detecting the aging degree of a rubber pad of a railway fastener, which are used to solve the technical problem of being unable to identify early damage and invisible defects of a structure.

[0005] In a first aspect, the present invention provides a method for detecting the aging degree of a rubber pad of a railway fastener, comprising:

[0006] Construct a refined model of fastener pads considering different temperature characteristics;

[0007] According to the fastener rubber pad refined model, the fastener rubber pad is simulated considering nonlinear effects, track unevenness and different fastener rubber pad aging degrees to obtain rail vibration acceleration in the time domain, and an original data set of fastener rubber pad aging is constructed according to the rail vibration acceleration;

[0008] Adopting adaptive frequency bandwidth and weight adjustment empirical wavelet transform to decompose the vibration response signal of the rail to obtain modal components, and classifying the fastener rubber pad aging degree according to the fastener rubber pad aging degree index of the modal components to obtain the fastener rubber pad aging degree classification result corresponding to the modal components;

[0009] Taking the modal component as input and the classification result of the fastener rubber pad aging degree as output, a convolutional neural network is trained to obtain a fastener rubber pad aging degree detection model;

[0010] The rail vibration acceleration data in the actual railway line is collected, and the rail vibration acceleration data is input into the fastener rubber pad aging degree detection model, and the fastener rubber pad aging degree detection model outputs the fastener rubber pad aging degree corresponding to the rail vibration acceleration data.

[0011] In a second aspect, the present invention provides a railway fastener rubber pad aging degree detection system, comprising:

[0012] A construction module configured to construct a refined model of a fastener pad considering different temperature characteristics;

[0013] a simulation module configured to simulate the fastener pads according to the fastener pad refinement model, taking into account nonlinear effects, track irregularities and different degrees of fastener pad aging, obtain rail vibration acceleration in the time domain, and construct an original data set of fastener pad aging according to the rail vibration acceleration;

[0014] A decomposition module is configured to decompose the vibration response signal of the rail by using an adaptive frequency bandwidth and weight-adjusted empirical wavelet transform to obtain modal components, and classify the fastener rubber pad aging degree according to the fastener rubber pad aging degree index to obtain a fastener rubber pad aging degree classification result corresponding to the modal component;

[0015] A training module is configured to take the modal component as input and the fastener rubber pad aging degree classification result as output, train a convolutional neural network, and obtain a fastener rubber pad aging degree detection model;

[0016] The output module is configured to collect rail vibration acceleration data in an actual railway line, input the rail vibration acceleration data into the fastener rubber pad aging degree detection model, and the fastener rubber pad aging degree detection model outputs the fastener rubber pad aging degree corresponding to the rail vibration acceleration data.

[0017] In a third aspect, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the steps of the method for detecting the aging degree of railway fastener rubber pads according to any embodiment of the present invention.

[0018] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor executes the steps of the method for detecting the aging degree of railway fastener rubber pads of any embodiment of the present invention.

[0019] The railway fastener rubber pad aging degree detection method and system of the present application establishes a fastener rubber pad refined model considering different temperature characteristics, simulates the fastener rubber pad aging degree by adjusting the rubber pad stiffness, obtains the rail vibration acceleration that can reflect the rubber pad aging degree, and uses adaptive frequency bandwidth and weight adjustment empirical wavelet transform to decompose the acceleration signal to obtain the modal component with damage characteristics, constructs a one-dimensional convolutional neural network based on the modal component and trains it, and uses the Adam optimization algorithm based on Frank-Wolfe to improve the fastener rubber pad aging degree detection efficiency of the convolutional neural network, which has the following advantages:

[0020] Different from the previous image-based defect detection methods, images are mainly limited to visible damage. In order to solve this limitation, a convolutional network is used to identify the aging degree of fastener pads and detect invisible aging damage of fastener pads.

[0021] The refined model of fastener pads was considered to conduct train-track coupling dynamics calculation analysis and simulation to build a large data set of track fastener pads for training CNN, solving the problem of difficulty in obtaining fastener pad aging data;

[0022] Since the adaptive frequency bandwidth and weight-adjusted empirical wavelet transform are used to extract the modal components sensitive to the aging degree of the fastener pad from the original rail vibration acceleration data as a data set, the neural network detection efficiency is higher, so compared with the traditional intelligent detection method and system, it has higher detection accuracy and speed;

[0023] The accuracy of defect detection of rubber pads of railway fasteners can be improved, and the robustness is strong. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0025] Figure 1 A flow chart of a method for detecting the aging degree of a rubber pad of a railway fastener provided by an embodiment of the present invention;

[0026] Figure 2 A structural block diagram of a railway fastener rubber pad aging degree detection system provided by an embodiment of the present invention;

[0027] Figure 3 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] See also Figure 1 , which shows a flow chart of a method for detecting the aging degree of a railway fastener rubber pad of the present application.

[0030] like Figure 1 As shown, the method for detecting the aging degree of the rubber pad of the railway fastener specifically includes the following steps:

[0031] Step S101, constructing a fastener pad refinement model taking into account different temperature characteristics.

[0032] In this step, the fastener pad model is refined to represent the aging degree of the fastener pad:

[0033] The stiffness of the fastener system is determined by the stiffness of the fastener Fastener pad stiffness Composition, among which the stiffness of the buckle Only about the stiffness of the fastener pad Therefore, the rigidity of the fastener pad It can be approximately regarded as the total stiffness of the entire fastener system.

[0034] At present, the calculation of the fastener rubber pad stiffness used in the vehicle-track coupling dynamics calculation does not consider the influence of temperature characteristics in the model, which is inaccurate. Therefore, the model of the present invention adopts an improved fastener rubber pad model to simulate the response caused by the aging degree of the fastener rubber pad that is more in line with reality. The details are as follows:

[0035] Fastener pad stiffness It can be expressed as: ;

[0036] In the formula, is the storage modulus of the fastener pad model, is the width of the fastener pad;

[0037] Storage modulus of the fastener pad refinement model considering different temperature characteristics It is obtained from the constitutive equation, and the expression of the constitutive equation is:

[0038] ,

[0039] In the formula, Fractional strain in the pad material The differential operator of is the differential operator describing the elastic behavior of the rubber pad material, is the differential operator describing the viscous behavior of the material, is the composite fractional derivative differential operator, is the displacement of the fastener pad, is the temperature influence function, The parameters of the influence of temperature on material properties obtained by the least squares method are: is the ambient temperature of the fastener pad. For normal temperature, take 20℃, The elastic force of the elastic cushion;

[0040] in, ,

[0041] ,

[0042] ,

[0043] In the formula, , , Indicators that indicate the relative strength of different elastic and viscous behaviors in the rubber pad material. , are the low elastic coefficient and high elastic coefficient of the rubber pad model respectively. , are the low viscosity coefficient and high viscosity coefficient of the rubber pad model respectively;

[0044] The constitutive equation is transformed and separated to obtain the fastener pad stiffness of the fastener pad refined model. The expression is:

[0045] ,

[0046] In the formula, is the fractional derivative order of an elastic behavior of the rubber pad material, is the fractional derivative order of another elastic behavior of the pad material, is the viscosity fractional derivative order of the pad material, is the circular frequency of the excitation, is the width of the fastener pad.

[0047] By changing the elastic coefficient in the model and viscosity coefficient Ratio To change the model's pad stiffness This reflects the actual aging condition of the fastener rubber pad.

[0048] Step S102, based on the fastener pad refined model, the fastener pad is simulated considering nonlinear effects, track unevenness and different fastener pad aging degrees to obtain rail vibration acceleration in the time domain, and an original data set of fastener pad aging is constructed based on the rail vibration acceleration.

[0049] In this step, random vibration response simulation excitation is used to solve the problem. In reality, the rail acceleration is affected not only by the stiffness of the fastener rubber pad, but also by the track unevenness. Therefore, the track unevenness spectral density is considered to be When the train is running, due to the different positions of the four wheels and rails, the excitation time between different wheels and rails will also be out of sync. Therefore, the dynamic wheel-rail force , the expression is:

[0050] ,

[0051] In the formula, is the wheel response at the wheel-rail contact point, is the response of the rail at the wheel-rail contact point, is the wheel-rail contact response, is the track irregularity spectral density, is the circular frequency of the excitation, is the center distance of the first pair of wheel rails, is the center distance of the second pair of wheels and rails, is the wheel volume;

[0052] The fastener pad stiffness , Dynamic wheel-rail force Combining the multi-degree-of-freedom system, frequency-dependent damping, and nonlinear processing, the rail vibration acceleration in the time domain is obtained according to the superposition principle of the vibration response of the wheel-rail coupling system. , the expression is:

[0053] ,

[0054] ,

[0055] In the formula, is the system mass matrix, is an imaginary unit, is the system damping matrix, is the system stiffness matrix, is the nonlinear model of temperature dependence, is the rail displacement response, is the ambient temperature of the fastener pad. is the temperature matrix function used to describe the effect of temperature on the strength of the nonlinear model. is the correlation coefficient between the model and the pad stiffness, is the correlation coefficient between the model and temperature, is the interaction coefficient between the model and temperature change, represents function composition, The temperature-dependent stiffness matrix is ​​used to describe the temperature dependence of the system. is the basic stiffness coefficient in the stiffness matrix, is the stiffness coefficient that changes with temperature, represents the Hadamard product, is the temperature-dependent nonlinear exponential vector used to indicate how the intensity of the nonlinear behavior in the corresponding dimension changes with temperature. is the baseline level correlation coefficient between the model and the nonlinear behavior, is the nonlinear coefficient of the model that changes with temperature. Step S103, the vibration response signal of the rail is decomposed to obtain modal components by using adaptive frequency bandwidth and weight-adjusted empirical wavelet transform, and the modal components are classified according to the fastener pad aging degree index to obtain the fastener pad aging degree classification result corresponding to the modal components.

[0056] In this step, the expression of the fastener rubber pad aging index is:

[0057] ,

[0058] In the formula, is the stiffness of the fastener pad, Design initial stiffness for fastener pads. Design initial stiffness for fastener pads The vibration acceleration of the rail under the action of is the circular frequency of the excitation, Dynamic wheel-rail forces, is the system mass matrix, is an imaginary unit, is the system damping matrix, is the system stiffness matrix, is the nonlinear model of temperature dependence, is the rail displacement response, It is the ambient temperature of the fastener pad.

[0059] The rubber pad aging evaluation index G has a value range of 0 to 1, corresponding to different aging degrees. G = 0 (ie, aging degree 0) corresponds to a healthy state, and G = 1 (ie, aging degree 1) corresponds to a failure state and needs to be replaced.

[0060] This allows the elastic coefficient in the model to be adjusted and viscosity coefficient Ratio To adjust the stiffness of the model fastener pad Thus, a large data set is obtained to simulate the aging degree of the fastener rubber pads, and the aging degree is classified according to the aging degree index G of the fastener rubber pads.

[0061] It should be noted that the original data set of fastener rubber pad aging is transformed into frequency domain data by Fourier transform, and then the acceleration frequency domain signal is adaptively divided by Fourier spectrum, where the expression of adaptive division is:

[0062] ,

[0063] ,

[0064] In the formula, is the transition segment of the nth spectrum partition width, The frequency magnitude of the original data set of fastener pad aging converted from Fourier transform, is the center frequency of the nth mode, For frequency The energy density function at is the frequency interval used to calculate The average power near is the Fourier transform of the acceleration signal, To adjust the sensitivity coefficient, is the bandwidth impact proportionality coefficient;

[0065] After the interval is determined, the empirical wavelet function is defined in the frequency domain according to the wavelet construction method. and the empirical scaling function , the expression is:

[0066] ,

[0067] ,

[0068] In the formula, is the transition segment of the nth spectrum partition width, , A transition segment for the width of the n+1th spectrum division;

[0069] The detailed classification of the sensitive weights for characterizing the effect of the reconstructed acceleration signal on the aging of the fastener pad is expressed as:

[0070] ,

[0071] ,

[0072] In the formula, is the frequency sensitive weight function, is the center frequency of the nth mode, exp(x) is the exponential function, is the pre-determined sensitive frequency associated with the aging of the rubber pad. To control the standard deviation of the weight decay rate, is the sinusoidal modulation depth coefficient, is the frequency parameter of the sine function, is the detail coefficient of the reconstructed signal, is the inner product operation, is the normalized data set, for The complex conjugate function of is the inverse Fourier transform;

[0073] Use convolution operation to get the modal component function , the expression is:

[0074] ,

[0075] In the formula, For the modal components, is the convolution operation, is the wavelet function;

[0076] The vibration response signal of the rail is decomposed into modal components by using adaptive frequency bandwidth and weight adjustment empirical wavelet transform, and the modal components are classified according to the fastener rubber pad aging degree index.

[0077] Step S104, taking the modal component as input and the fastener rubber pad aging degree classification result as output, training a convolutional neural network to obtain a fastener rubber pad aging degree detection model.

[0078] In this step, the Frank-Wolfe-based Adam optimization algorithm is used to perform exponential weighted optimization on the batch stochastic gradient for network training.

[0079] Step S105, collecting rail vibration acceleration data in the actual railway line, inputting the rail vibration acceleration data into the fastener rubber pad aging degree detection model, and the fastener rubber pad aging degree detection model outputs the fastener rubber pad aging degree corresponding to the rail vibration acceleration data.

[0080] In summary, the method of the present application, based on the vehicle-track coupling dynamic analysis model, simulates the aging degree of the fastener rubber pads by increasing the stiffness of the fastener rubber pads as they age; the simulation considers nonlinear effects, track unevenness and different degrees of aging of fastener rubber pads to obtain rail vibration acceleration, and then uses adaptive frequency bandwidth and weight adjustment empirical wavelet transform to decompose the vibration response signal of the rail to obtain modal components to construct a large data set. A one-dimensional convolutional neural network is designed, and the established network is trained, cross-validated and optimized using the above data set. The trained network is tested for performance on the test set, and the test results show that the detection method and system have high detection accuracy and robustness.

[0081] See also Figure 2 , which shows a structural block diagram of a railway fastener rubber pad aging degree detection system of the present application.

[0082] like Figure 2 As shown, the railway fastener rubber pad aging degree detection system 200 includes a construction module 210, a simulation module 220, a decomposition module 230, a training module 240 and an output module 250.

[0083] Among them, the construction module 210 is configured to construct a refined model of fastener pads considering different temperature characteristics; the simulation module 220 is configured to simulate the fastener pads according to the refined model of the fastener pads, considering nonlinear effects, track unevenness and different degrees of aging of the fastener pads, to obtain the rail vibration acceleration in the time domain, and to construct the original data set of fastener pad aging according to the rail vibration acceleration; the decomposition module 230 is configured to decompose the vibration response signal of the rail using adaptive frequency bandwidth and weight adjustment empirical wavelet transform to obtain modal components, and to classify the modal components according to the fastener pad aging degree index. The fastener rubber pad aging degree is classified to obtain the classification result of the fastener rubber pad aging degree corresponding to the modal component; the training module 240 is configured to take the modal component as input and the classification result of the fastener rubber pad aging degree as output, train the convolutional neural network to obtain a fastener rubber pad aging degree detection model; the output module 250 is configured to collect rail vibration acceleration data in the actual railway line, input the rail vibration acceleration data into the fastener rubber pad aging degree detection model, and the fastener rubber pad aging degree detection model outputs the fastener rubber pad aging degree corresponding to the rail vibration acceleration data.

[0084] It should be understood that Figure 2 Modules and references documented in Figure 1 Therefore, the operations and features described above for the method and the corresponding technical effects are also applicable to Figure 2 The modules in it will not be described in detail here.

[0085] In some other embodiments, the embodiments of the present invention further provide a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor is caused to execute the method for detecting the aging degree of the rubber pad of the railway fastener in any of the above method embodiments;

[0086] As an implementation mode, the computer-readable storage medium of the present invention stores computer-executable instructions, and the computer-executable instructions are configured as follows:

[0087] Construct a refined model of fastener pads considering different temperature characteristics;

[0088] According to the fastener rubber pad refined model, the fastener rubber pad is simulated considering nonlinear effects, track unevenness and different fastener rubber pad aging degrees to obtain rail vibration acceleration in the time domain, and an original data set of fastener rubber pad aging is constructed according to the rail vibration acceleration;

[0089] Adopting adaptive frequency bandwidth and weight adjustment empirical wavelet transform to decompose the vibration response signal of the rail to obtain modal components, and classifying the fastener rubber pad aging degree according to the fastener rubber pad aging degree index of the modal components to obtain the fastener rubber pad aging degree classification result corresponding to the modal components;

[0090] Taking the modal component as input and the classification result of the fastener rubber pad aging degree as output, a convolutional neural network is trained to obtain a fastener rubber pad aging degree detection model;

[0091] The rail vibration acceleration data in the actual railway line is collected, and the rail vibration acceleration data is input into the fastener rubber pad aging degree detection model, and the fastener rubber pad aging degree detection model outputs the fastener rubber pad aging degree corresponding to the rail vibration acceleration data.

[0092] The computer-readable storage medium may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created according to the use of the railway fastener rubber pad aging degree detection system, etc. In addition, the computer-readable storage medium may include a high-speed random access memory, and may also include a memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the railway fastener rubber pad aging degree detection system via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0093] Figure 3is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 3 As shown, the device includes: a processor 310 and a memory 320. The electronic device may also include: an input device 330 and an output device 340. The processor 310, the memory 320, the input device 330 and the output device 340 may be connected via a bus or other means. Figure 3 In the example, the connection through the bus is taken as an example. The memory 320 is the above-mentioned computer-readable storage medium. The processor 310 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions and modules stored in the memory 320, that is, the railway fastener rubber pad aging degree detection method of the above-mentioned method embodiment is realized. The input device 330 can receive input digital or character information, and generate key signal input related to the user setting and function control of the railway fastener rubber pad aging degree detection system. The output device 340 may include a display device such as a display screen.

[0094] The electronic device can execute the method provided by the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not described in detail in this embodiment, please refer to the method provided by the embodiment of the present invention.

[0095] As an implementation mode, the electronic device is applied to a railway fastener rubber pad aging degree detection system, and is used for a client, and includes: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can:

[0096] Construct a refined model of fastener pads considering different temperature characteristics;

[0097] According to the fastener rubber pad refined model, the fastener rubber pad is simulated considering nonlinear effects, track unevenness and different fastener rubber pad aging degrees to obtain rail vibration acceleration in the time domain, and an original data set of fastener rubber pad aging is constructed according to the rail vibration acceleration;

[0098] Adopting adaptive frequency bandwidth and weight adjustment empirical wavelet transform to decompose the vibration response signal of the rail to obtain modal components, and classifying the fastener rubber pad aging degree according to the fastener rubber pad aging degree index of the modal components to obtain the fastener rubber pad aging degree classification result corresponding to the modal components;

[0099] Taking the modal component as input and the classification result of the fastener rubber pad aging degree as output, a convolutional neural network is trained to obtain a fastener rubber pad aging degree detection model;

[0100] The rail vibration acceleration data in the actual railway line is collected, and the rail vibration acceleration data is input into the fastener rubber pad aging degree detection model, and the fastener rubber pad aging degree detection model outputs the fastener rubber pad aging degree corresponding to the rail vibration acceleration data.

[0101] Through the description of the above implementation modes, those skilled in the art can clearly understand that each implementation mode can be implemented by means of software plus a necessary general hardware platform, or of course by hardware. Based on such an understanding, the above technical solution can essentially or in other words be embodied in the form of a software product that contributes to the prior art. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiment.

[0102] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting the aging degree of a railway fastener rubber pad, characterized in that: include: Construct a refined model of fastener pads considering different temperature characteristics; According to the fastener rubber pad refined model, the fastener rubber pad is simulated considering nonlinear effects, track unevenness and different fastener rubber pad aging degrees to obtain rail vibration acceleration in the time domain, and an original data set of fastener rubber pad aging is constructed according to the rail vibration acceleration; Adopting adaptive frequency bandwidth and weight adjustment empirical wavelet transform to decompose the vibration response signal of the rail to obtain modal components, and classifying the fastener rubber pad aging degree according to the fastener rubber pad aging degree index of the modal components to obtain the fastener rubber pad aging degree classification result corresponding to the modal components; Taking the modal component as input and the classification result of the fastener rubber pad aging degree as output, a convolutional neural network is trained to obtain a fastener rubber pad aging degree detection model; The rail vibration acceleration data in the actual railway line is collected, and the rail vibration acceleration data is input into the fastener rubber pad aging degree detection model, and the fastener rubber pad aging degree detection model outputs the fastener rubber pad aging degree corresponding to the rail vibration acceleration data.

2. A method for detecting the aging degree of railway fastener rubber pads according to claim 1, characterized in that: The construction of a fastener pad refinement model considering different temperature characteristics includes: Storage modulus of the fastener pad refinement model considering different temperature characteristics It is obtained from the constitutive equation, and the expression of the constitutive equation is: ， In the formula, Fractional strain in the pad material The differential operator of is the differential operator describing the elastic behavior of the rubber pad material, is the differential operator describing the viscous behavior of the material, is the composite fractional derivative differential operator, is the displacement of the fastener pad, is the temperature influence function, The parameters of the influence of temperature on material properties obtained by the least squares method are: is the ambient temperature of the fastener pad. For normal temperature, take 20℃, The elastic force of the elastic cushion; in, , ， ， In the formula, , , Indicators that indicate the relative strength of different elastic and viscous behaviors in the rubber pad material. , are the low elastic coefficient and high elastic coefficient of the rubber pad model respectively. , are the low viscosity coefficient and high viscosity coefficient of the rubber pad model respectively; The constitutive equation is transformed and separated to obtain the fastener pad stiffness of the fastener pad refined model The expression is: ， In the formula, is the fractional derivative order of an elastic behavior of the rubber pad material, is the fractional derivative order of another elastic behavior of the pad material, is the viscosity fractional derivative order of the pad material, is the circular frequency of the excitation, is the width of the fastener pad.

3. A method for detecting the aging degree of railway fastener rubber pads according to claim 1, characterized in that: According to the fastener rubber pad refined model, the fastener rubber pad is simulated considering nonlinear effects, track unevenness and different fastener rubber pad aging degrees, and the rail vibration acceleration in the time domain is obtained, including: When the train is running, due to the different wheel-rail positions, the excitation time between different wheels and rails will also be out of sync, resulting in the dynamic wheel-rail force , the expression is: ， In the formula, is the wheel response at the wheel-rail contact point, is the response of the rail at the wheel-rail contact point, is the wheel-rail contact response, is the track irregularity spectral density, is the circular frequency of the excitation, is the center distance of the first pair of wheel rails, is the center distance of the second pair of wheels and rails, is the wheel volume; The fastener pad stiffness , Dynamic wheel-rail force Combining the multi-degree-of-freedom system, frequency-dependent damping, and nonlinear processing, according to the superposition principle of the vibration response of the wheel-rail coupling system, the rail vibration acceleration in the time domain is obtained. , the expression is: ， ， In the formula, is the system mass matrix, is an imaginary unit, is the system damping matrix, is the system stiffness matrix, is the nonlinear model of temperature dependence, is the rail displacement response, is the ambient temperature of the fastener pad. is the temperature matrix function used to describe the effect of temperature on the strength of the nonlinear model. is the correlation coefficient between the model and the pad stiffness, is the correlation coefficient between the model and temperature, is the interaction coefficient between the model and temperature change, represents function composition, The temperature-dependent stiffness matrix is ​​used to describe the temperature dependence of the system. is the basic stiffness coefficient in the stiffness matrix, is the stiffness coefficient that changes with temperature, represents the Hadamard product, is the temperature-dependent nonlinear exponential vector used to indicate how the intensity of the nonlinear behavior in the corresponding dimension changes with temperature. is the baseline level correlation coefficient between the model and the nonlinear behavior, is the nonlinear coefficient of the model that changes with temperature.

4. A method for detecting the aging degree of a railway fastener rubber pad according to claim 1, characterized in that: The expression of the fastener rubber pad aging index is: ， In the formula, is the stiffness of the fastener pad, Design initial stiffness for fastener pads. Design initial stiffness for fastener pads The vibration acceleration of the rail under the action of is the circular frequency of the excitation, Dynamic wheel-rail forces, is the system mass matrix, is an imaginary unit, is the system damping matrix, is the system stiffness matrix, is the nonlinear model of temperature dependence, is the rail displacement response, It is the ambient temperature of the fastener pad.

5. A method for detecting the aging degree of railway fastener rubber pads according to claim 1, characterized in that: The method of using adaptive frequency bandwidth and weight adjustment empirical wavelet transform to decompose the vibration response signal of the rail to obtain modal components, and classifying the aging degree of the fastener rubber pad according to the fastener rubber pad aging degree index includes: The original data set of fastener rubber pad aging is transformed into frequency domain data by Fourier transform, and then the acceleration frequency domain signal is adaptively divided by Fourier spectrum. The expression of adaptive division is: ， ， In the formula, is the transition segment of the nth spectrum partition width, The frequency magnitude of the original data set of fastener pad aging converted from Fourier transform, is the center frequency of the nth mode, For frequency The energy density function at is the frequency interval used to calculate The average power near is the Fourier transform of the acceleration signal, To adjust the sensitivity coefficient, is the bandwidth impact proportionality coefficient; After the interval is determined, the empirical wavelet function is defined in the frequency domain according to the wavelet construction method. and the empirical scaling function , the expression is: ， ， In the formula, is the transition segment of the nth spectrum partition width, , A transition segment for the width of the n+1th spectrum division; The detailed classification of the sensitive weights for characterizing the effect of the reconstructed acceleration signal on the aging of the fastener pad is expressed as: ， ， In the formula, is the frequency sensitive weight function, is the center frequency of the nth mode, exp(x) is the exponential function, is the pre-determined sensitive frequency associated with the aging of the rubber pad. To control the standard deviation of the weight decay rate, is the sinusoidal modulation depth coefficient, is the frequency parameter of the sine function, is the detail coefficient of the reconstructed signal, is the inner product operation, is the normalized data set, for The complex conjugate function of is the inverse Fourier transform; Use convolution operation to get the modal component function , the expression is: ， In the formula, For the modal components, is the convolution operation, is the wavelet function; The vibration response signal of the rail is decomposed into modal components by using adaptive frequency bandwidth and weight adjustment empirical wavelet transform, and the modal components are classified according to the fastener rubber pad aging degree index.

6. A railway fastener rubber pad aging degree detection system, characterized in that: include: A construction module configured to construct a refined model of a fastener pad considering different temperature characteristics; a simulation module configured to simulate the fastener pads according to the fastener pad refinement model, taking into account nonlinear effects, track irregularities and different degrees of fastener pad aging, obtain rail vibration acceleration in the time domain, and construct an original data set of fastener pad aging according to the rail vibration acceleration; A decomposition module is configured to decompose the vibration response signal of the rail by using an adaptive frequency bandwidth and weight-adjusted empirical wavelet transform to obtain modal components, and classify the fastener rubber pad aging degree according to the fastener rubber pad aging degree index to obtain a fastener rubber pad aging degree classification result corresponding to the modal component; A training module is configured to take the modal component as input and the fastener rubber pad aging degree classification result as output, train a convolutional neural network, and obtain a fastener rubber pad aging degree detection model; The output module is configured to collect rail vibration acceleration data in an actual railway line, input the rail vibration acceleration data into the fastener rubber pad aging degree detection model, and the fastener rubber pad aging degree detection model outputs the fastener rubber pad aging degree corresponding to the rail vibration acceleration data.

7. An electronic device, characterized in that: include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method described in any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images