Metro rail corrugation identification method and system based on vehicle vibration and noise combined test

Abstract

The present invention provides a subway rail corrugation identification method and system based on vehicle vibration and noise combined testing, which belongs to the field of rail transit technology, and obtains train operation data; calculates the relationship between the train body vibration acceleration and the noise inside the car and the mileage according to the train operation data; divides the body vibration acceleration and the noise inside the car into unit sections according to the mileage, and obtains the body acceleration and noise inside the car characteristics of each section; uses a pre-trained recognition model to process the body vibration acceleration and noise inside the car data characteristics to realize rail corrugation identification; realizes the identification of the wavelength and amplitude of rail corrugation according to the vehicle speed, body vibration acceleration and the frequency and amplitude of the noise inside the car. The present invention improves the detection efficiency and accuracy of rail corrugation, can easily and quickly identify rail corrugation and its specific characteristics, saves the maintenance and repair costs of subway tracks, and is of great significance for ensuring the safe and stable operation of subway vehicles and guiding track maintenance and repair.

Description

Technical Field

[0001] The present invention relates to the field of rail transit technology, and in particular to a subway rail corrugation identification method and system based on vehicle vibration and noise combined testing. Background Art

[0002] After a period of operation, rail corrugation will appear on the top surface of the rails along the driving direction. Rail corrugation will cause intense vibrations between the vehicle and the track structure, significantly shortening the service life of various line components. In addition, subway trains will generate severe in-car noise when passing through the rail corrugation section, affecting the comfort of train operation. Therefore, how to accurately and efficiently identify rail corrugation is an important issue that needs to be solved urgently.

[0003] In terms of rail corrugation detection, the subway operation and maintenance department mainly uses a variety of equipment such as dynamic inspection vehicles and flaw detection vehicles to conduct periodic inspections on track structures, which can be divided into hand-push and vehicle-mounted types. The hand-push rail corrugation detection method is mainly based on the acceleration integration method and the chord measurement method. The principle is complex, the processing accuracy is high, and the purchase and maintenance costs are also high, such as the CAT corrugation measurement trolley (imported single equipment is more than 500,000 yuan). The vehicle-mounted rail corrugation detection method is mainly based on the vehicle axle box acceleration to diagnose rail corrugation. The acceleration sensor needs to be arranged at the position of the operating vehicle axle box, and a lot of cost needs to be invested in daily maintenance. It is difficult to achieve full coverage of sensors on the existing operating vehicle axle boxes, and the cost is huge. At present, there are also methods for detecting rail corrugation based on vehicle body acceleration, but they can only qualitatively detect rail corrugation and cannot accurately identify the specific characteristics of corrugation. Summary of the invention

[0004] The purpose of the present invention is to provide a method and system for identifying subway rail corrugation based on vehicle vibration and noise combined testing, which improves the efficiency and accuracy of rail corrugation detection and provides important data support for subway operation and maintenance departments to formulate scientific and efficient maintenance strategies, so as to solve at least one technical problem existing in the above-mentioned background technology.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] In one aspect, the present invention provides a method for identifying subway rail corrugation based on vehicle vibration and noise combined testing, comprising:

[0007] Acquire train operation data; wherein the operation data includes vehicle body vibration acceleration, in-vehicle noise data, operation time, speed and mileage information during operation;

[0008] Calculate the relationship between the car body vibration acceleration and the noise inside the car and the mileage during the train operation process based on the train operation data;

[0009] The vehicle body vibration acceleration and interior noise are divided into unit sections according to mileage to obtain the characteristics of vehicle body acceleration and interior noise in each section;

[0010] The pre-trained recognition model is used to process the data features of vehicle body vibration acceleration and in-vehicle noise to realize the recognition of rail corrugation; the wavelength and amplitude of rail corrugation are recognized based on the vehicle speed, vehicle body vibration acceleration and the frequency and amplitude of in-vehicle noise.

[0011] Preferably, based on the changing relationship between the body vibration acceleration and the in-car noise data over time, combined with the changing relationship between the train running mileage over time, time is regarded as an intermediate variable, thereby obtaining the changing relationship between the body vibration acceleration and the in-car noise over mileage.

[0012] Preferably, obtaining the vehicle body acceleration and interior noise characteristics of each section includes: dividing the vehicle body vibration acceleration data and the interior noise data into units, and extracting the maximum value, minimum value, peak-to-peak value, kurtosis, effective value, peak factor, margin factor, and pulse factor from the vehicle body acceleration and interior noise data of each unit section.

[0013] Preferably, training the recognition model includes: using the data features of each divided unit section as a training set, selecting a radial basis kernel function as a kernel function of a support vector machine, introducing the radial basis kernel function into the support vector machine, and obtaining the optimal classification function of the support vector machine; using a grid method to optimize the penalty factor and the width of the radial basis of the support vector machine; calculating the classification accuracy, and obtaining the penalty factor and the width of the radial basis when the classification accuracy is the highest from the qualified classification accuracy as the optimal parameters, and constructing an optimal support vector machine rail corrugation recognition model.

[0014] Preferably, the train speed of the identified rail corrugation section is calculated, the main frequency of the vehicle body vibration acceleration and the main frequency of the interior noise of the rail corrugation section are calculated using the Fourier transform method, and the common frequency of the main frequency of the vehicle body vibration acceleration and the main frequency of the interior noise is taken as the rail corrugation passing frequency; the characteristic wavelength of the rail corrugation is calculated based on the train speed of the identified rail corrugation section and the rail corrugation passing frequency, so as to realize the identification of the rail corrugation wavelength.

[0015] Preferably, the Fourier transform method is used to calculate the main frequency amplitude of the vehicle body vibration acceleration and the main frequency amplitude of the interior noise in the rail corrugation section, and the regression analysis method is used to establish a mapping relationship between the main frequency amplitude of the vehicle body vibration acceleration, the main frequency amplitude of the interior noise and the rail corrugation amplitude, so as to realize the identification of the rail corrugation amplitude.

[0016] In a second aspect, the present invention provides a subway rail corrugation identification system based on vehicle vibration and noise combined testing, comprising:

[0017] An acquisition module is used to acquire train operation data; wherein the operation data includes vehicle body vibration acceleration, in-vehicle noise data, operation time, speed and mileage information during operation;

[0018] A calculation module, used to calculate the relationship between the car body vibration acceleration and the noise inside the car and the mileage during the train operation process according to the train operation data;

[0019] A division module is used to divide the vehicle body vibration acceleration and the interior noise into unit sections according to mileage, and obtain the vehicle body acceleration and interior noise characteristics of each section;

[0020] The recognition module is used to process the characteristics of vehicle body vibration acceleration and in-vehicle noise data using a pre-trained recognition model to realize the recognition of rail corrugation; the wavelength and amplitude of rail corrugation are recognized according to the vehicle speed, vehicle body vibration acceleration and the frequency and amplitude of in-vehicle noise.

[0021] In a third aspect, the present invention provides a computer program product, including a computer program, which, when executed on one or more processors, is used to implement the subway rail corrugation identification method based on vehicle vibration and noise combined testing as described above.

[0022] In a fourth aspect, the present invention provides a non-transitory computer-readable storage medium, which is used to store computer instructions. When the computer instructions are executed by a processor, the subway rail corrugation identification method based on vehicle vibration and noise combined testing as described above is implemented.

[0023] In a fifth aspect, the present invention provides an electronic device, comprising: a processor, a memory and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory so that the electronic device executes instructions for implementing the subway rail corrugation identification method based on vehicle vibration and noise combined testing as described above.

[0024] Beneficial effects of the present invention:

[0025] The vibration acceleration of the vehicle body and the noise inside the vehicle, as well as their changing relationship with mileage, were obtained by using acceleration sensors and sound pressure sensors; the vibration and noise data were segmented, and eight time domain features of the vehicle body vibration acceleration and eight time domain features of the noise inside the vehicle were extracted in each segment; the data features were classified with the help of the support vector machine method to realize the identification of subway rail corrugation, and the wavelength and amplitude of the rail corrugation were further determined based on the vibration and noise frequency domain characteristics; the detection efficiency and accuracy of rail corrugation were greatly improved, and rail corrugation and its specific characteristics can be easily and quickly identified, saving the maintenance cost of subway tracks, which is of great significance for ensuring the safe and smooth operation of subway vehicles and guiding track maintenance.

[0026] Additional aspects and advantages of the present invention will be given in part in the following description, which will become obvious from the following description, or may be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0028] Figure 1 The present invention is a flowchart of a method for identifying subway rail corrugation based on vehicle vibration and noise combined testing according to an embodiment of the present invention.

[0029] Figure 2 This is a flow chart of rail corrugation recognition using a support vector machine according to an embodiment of the present invention. DETAILED DESCRIPTION

[0030] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The embodiments described below by the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be interpreted as limiting the present invention.

[0031] It should be understood by those skilled in the art that unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those skilled in the art to which this invention belongs.

[0032] It should also be understood that terms, such as those defined in commonly used dictionaries, should be understood to have a meaning consistent with that in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless as defined herein.

[0033] Those skilled in the art will appreciate that, unless otherwise stated, the singular forms "a", "an", "said" and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements and / or groups thereof.

[0034] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine different embodiments or examples described in this specification and the features of different embodiments or examples, unless they are contradictory.

[0035] To facilitate understanding of the present invention, the present invention is further explained below with reference to specific embodiments in conjunction with the accompanying drawings, and the specific embodiments do not constitute a limitation on the embodiments of the present invention.

[0036] Those skilled in the art should understand that the drawings are merely schematic diagrams of embodiments, and the components in the drawings are not necessarily necessary for implementing the present invention.

[0037] Example 1

[0038] like Figure 1 As shown, this embodiment 1 provides a subway rail corrugation identification system based on vehicle vibration and noise combined testing, the system comprising:

[0039] An acquisition module is used to acquire train operation data; wherein the operation data includes vehicle body vibration acceleration, in-vehicle noise data, operation time, speed and mileage information during operation;

[0040] A calculation module, used to calculate the relationship between the car body vibration acceleration and the noise inside the car and the mileage during the train operation process according to the train operation data;

[0041] A division module is used to divide the vehicle body vibration acceleration and the interior noise into unit sections according to mileage, and obtain the vehicle body acceleration and interior noise characteristics of each section;

[0042] The recognition module is used to process the characteristics of vehicle body vibration acceleration and in-vehicle noise data using a pre-trained recognition model to realize the recognition of rail corrugation; the wavelength and amplitude of rail corrugation are recognized according to the vehicle speed, vehicle body vibration acceleration and the frequency and amplitude of in-vehicle noise.

[0043] In this embodiment 1, the above system is used to implement a subway rail corrugation identification method based on vehicle vibration and noise combined testing, including:

[0044] The train operation data is acquired by using an acquisition module; wherein the operation data includes the vehicle body vibration acceleration, vehicle interior noise data, operation time, speed and mileage information during operation;

[0045] The calculation module is used to calculate the relationship between the car body vibration acceleration and the noise inside the car during the train operation and the mileage according to the train operation data;

[0046] The vehicle body vibration acceleration and interior noise are divided into unit sections according to mileage using the division module to obtain the vehicle body acceleration and interior noise characteristics of each section;

[0047] Using the recognition module, based on the pre-trained recognition model, the data characteristics of the vehicle body vibration acceleration and the interior noise are processed to realize the recognition of rail corrugation; according to the vehicle speed, the vehicle body vibration acceleration and the frequency and amplitude of the interior noise, the wavelength and amplitude of the rail corrugation are recognized.

[0048] According to the changing relationship between the body vibration acceleration and the interior noise data with time, combined with the changing relationship between the train running mileage and time, time is regarded as an intermediate variable, thus obtaining the changing relationship between the body vibration acceleration and the interior noise with mileage.

[0049] In this embodiment 1, obtaining the vehicle body acceleration and interior noise characteristics of each section includes: dividing the vehicle body vibration acceleration data and the interior noise data into units, and extracting the maximum value, minimum value, peak-to-peak value, kurtosis, effective value, peak factor, margin factor, and pulse factor of the vehicle body acceleration and interior noise data of each unit section.

[0050] Training the recognition model includes: using the data features of each divided unit section as a training set, selecting a radial basis kernel function as a kernel function of a support vector machine, introducing the radial basis kernel function into the support vector machine, and obtaining an optimal classification function of the support vector machine; using a grid method to optimize the penalty factor of the support vector machine and the width of the radial basis; calculating the classification accuracy, and obtaining the penalty factor and the width of the radial basis when the classification accuracy is the highest from the qualified classification accuracy as optimal parameters, and constructing an optimal support vector machine rail corrugation recognition model.

[0051] The train speed of the identified rail corrugation section is calculated, and the main frequency of the car body vibration acceleration and the main frequency of the interior noise in the rail corrugation section are calculated using the Fourier transform method. The common frequency of the main frequency of the car body vibration acceleration and the main frequency of the interior noise is taken as the rail corrugation passing frequency. The characteristic wavelength of the rail corrugation is calculated based on the train speed of the identified rail corrugation section and the rail corrugation passing frequency, so as to realize the identification of the rail corrugation wavelength.

[0052] The Fourier transform method is used to calculate the main frequency amplitude of the car body vibration acceleration and the main frequency amplitude of the interior noise in the rail corrugation section. The regression analysis method is used to establish the mapping relationship between the main frequency amplitude of the car body vibration acceleration, the main frequency amplitude of the interior noise and the rail corrugation amplitude, so as to realize the identification of the rail corrugation amplitude.

[0053] Example 2

[0054] like Figure 1 As shown, this embodiment 2 provides a subway rail corrugation identification method based on vehicle vibration and noise combined testing, which specifically includes the following steps:

[0055] S1. Obtaining the vehicle body vibration acceleration and vehicle interior noise data during train operation;

[0056] S2, obtaining the train running time, speed and mileage information;

[0057] S3, obtaining the relationship between the car body vibration acceleration and the noise inside the car and the mileage during the train operation;

[0058] S4, dividing the vehicle body vibration acceleration and the vehicle interior noise into unit sections according to the mileage, and obtaining the vehicle body acceleration and vehicle interior noise characteristics of each section;

[0059] S5. Use support vector machine to train the characteristics of vehicle body vibration acceleration and in-vehicle noise data to identify rail corrugation;

[0060] S6. According to the vehicle speed, vehicle body vibration acceleration and the frequency and amplitude of the vehicle interior noise, the wavelength and amplitude of the rail corrugation are identified.

[0061] The specific method of S1 is:

[0062] Fix the acceleration sensor on the floor of the subway train compartment; fix the sound pressure sensor above the compartment floor through a tripod, and ensure that there are no obstructions in the circular area with a radius of 1m with the sound pressure sensor as the center. The sound pressure sensor should be calibrated before use.

[0063] The specific method of S2 is:

[0064] The longitudinal acceleration sensor is fixed on the floor of the subway train compartment to obtain the acceleration of the train during operation. τis the sensor sampling interval, X(t) is the train running distance, X′(t) is the train running speed, and X″(t) is the train running acceleration. In the time period [0, t], there are m unit sampling intervals. In each unit sampling interval [t q-1 ,t q ]Inside:

[0065] Train speed:

[0066] Train running distance:

[0067] Where X(t0)=0, X′(t0)=0

[0068] According to the above formula, the speed and mileage of the train at any time can be obtained.

[0069] The specific methods of S3 are:

[0070] The relationship between the body vibration acceleration and the noise inside the car and time is obtained in S1, and the relationship between the train mileage and time is obtained in S2. Therefore, time can be regarded as an intermediate variable to obtain the relationship between the body vibration acceleration and the noise inside the car and mileage.

[0071] The specific method of S4 is:

[0072] S4-1: Divide the vehicle body vibration acceleration data and the vehicle interior noise data into units, where the length of the division unit is L. It is assumed that the number of points per unit length of the vehicle body acceleration data is N1, and the number of points per unit length of the vehicle interior noise data is N2.

[0073] S4-2: Extract the maximum value, minimum value, peak-to-peak value, kurtosis, effective value, peak factor, margin factor, and pulse factor of the vehicle acceleration and interior noise data of each unit section;

[0074] Assume the acceleration of the vehicle body is a and the sound pressure inside the vehicle is p.

[0075] (1) Maximum value:

[0076] The maximum value of the vehicle acceleration is the maximum value of the N1 data points within the length range of L, expressed as a max Indicates that the maximum value of the noise inside the car is the maximum value of the N2 data points within the length range of L, expressed as p max express.

[0077] (2) Minimum value:

[0078] The minimum value of the vehicle acceleration is the minimum value among the N1 data points within the length range of L, expressed as a minIndicates that the minimum value of the noise inside the car is the minimum value of the N2 data points within the length range of L, expressed as p max express.

[0079] (3) Peak-to-peak value:

[0080] The peak-to-peak value describes the range of variation of the signal. The presence of corrugation increases the oscillation between the wheel and rail, which causes the peak-to-peak value of the signal to increase, that is, the range of variation of the signal's vibration amplitude increases.

[0081] The peak-to-peak value of the vehicle acceleration is the peak-to-peak value of N1 data points within the length range of L, expressed as a pp Means: a pp =a max -a min .

[0082] The peak-to-peak value of the noise inside the car is the peak-to-peak value of N2 data points within the length range of L, expressed as p pp Meaning: p pp =p max -p min .

[0083] (4) Kurtosis:

[0084] Kurtosis is the average value of the fourth-order moment of the signal, which can detect the impact component in the signal.

[0085] The kurtosis of vehicle acceleration is the kurtosis of N1 data points within the length range of L, expressed as K a express:

[0086]

[0087] The kurtosis of the noise inside the car is the kurtosis of N2 data points within the length range of L, expressed as K p express:

[0088]

[0089] Where μ is the mean of the signal.

[0090] (5) Effective value:

[0091] The root mean square value, also known as the effective value, is an indicator used to measure signal energy and is also an important indicator for determining the health status of the system in fault diagnosis.

[0092] The effective value of the vehicle acceleration is the effective value of N1 data points within the length range of L, expressed as express:

[0093] The effective value of the noise inside the car is the effective value of N2 data points within the length range of L, and is expressed as express:

[0094] (6) Peak factor:

[0095] The vehicle acceleration peak factor is the peak factor of N1 data points within the length range of L, expressed as express:

[0096] The peak factor of the noise inside the car is the peak factor of N2 data points within the length range of L, expressed as express:

[0097] (7) Margin factor:

[0098] The vehicle acceleration margin factor is the margin factor of N1 data points within the length range of L, expressed as express: in, is the square root amplitude.

[0099] The noise margin factor in the vehicle is the margin factor of N2 data points within the length range of L, expressed as express: in, is the square root amplitude.

[0100] (8) Pulse factor:

[0101] The vehicle acceleration pulse factor is the pulse factor of N1 data points within the length range of L, expressed as express: in, is the absolute average amplitude.

[0102] The noise pulse factor in the car is the pulse factor of N2 data points within the length range of L, expressed as express: in, is the absolute average amplitude.

[0103] The specific process of S5 is as follows Figure 2 As shown:

[0104] S5-1: Divide the data features of each unit segment divided in S4 into a training set and a test set.

[0105] S5-2: Based on the training set data, the slack variable ξn is introduced into the original optimization problem of the support vector machine to solve the generalized optimal hyperplane. The generalized optimal hyperplane is:

[0106]

[0107] The optimization conditions are:

[0108] y n ·(<ω·x n )+b)+ξ n -1≥0, n=1, 2, 3…N1+N2

[0109] ξ n ≥0, n=1, 2, 3..., N1+N2

[0110] C≥0

[0111] Among them, T(ω,ξ) is the optimization objective function, ω and b are the normal vector and intercept of the hyperplane respectively, and x n is the nth characteristic parameter of each unit segment in S4, N1+N2 represents the total number of data features of each unit segment in S4, and y n is the corresponding category identifier, y n = +1 indicates that the sample is a positive sample, y n =-1 means the sample is a negative sample, and C is called the penalty factor.

[0112] S5-3: The radial basis kernel function is selected as the kernel function of the support vector machine. The radial basis function is:

[0113] K(x n , x m ) = exp(-||x n -x m || 2 / 2g 2 )

[0114] Where g is the width of the radial basis; n, m = (1, 2, 3, ..., N1 + N2).

[0115] S5-4: Introduce the radial basis kernel function into the support vector machine, and obtain the optimal classification function of the support vector machine:

[0116]

[0117] Among them, sgn() is the symbol function, is the optimal Lagrangian coefficient, b * is the threshold value, x n and x m Indicates the number of sample points.

[0118] S5-5: Use the grid method to optimize the penalty factor C of the support vector machine and the width g of the radial basis; set the parameters C and g in the range of 2-10 to 210, take the iteration step size as 1, calculate the classification accuracy, and obtain the C and g with the highest classification accuracy from the qualified classification accuracy as the optimal parameters to construct the optimal support vector machine rail corrugation recognition model.

[0119] S5-6: For the test set data, the data features of each unit section are input into the trained support vector machine rail corrugation recognition model, and the accuracy index P is used to evaluate the recognition effect to realize the recognition of rail corrugation.

[0120]

[0121] TP represents the number of unit sections of rail corrugation that are correctly detected, and FP represents the number of unit sections of non-rail corrugation that are identified as rail corrugation.

[0122] The specific method of S6 is:

[0123] S6-1: using the method described in S2, calculate the train speed in the rail corrugation section identified in S5, denoted as v;

[0124] S6-2: The Fourier transform method is used to calculate the main frequency of the vehicle body vibration acceleration and the main frequency of the vehicle interior noise in the rail corrugation section. The common frequency of the two main frequencies is taken as the rail corrugation passing frequency f. The characteristic wavelength of corrugation is λ. According to the formula

[0125]

[0126] Calculate the characteristic wavelength of corrugation to identify the wavelength of rail corrugation.

[0127] S6-3: The Fourier transform method is used to calculate the main frequency amplitude of the car body vibration acceleration and the main frequency amplitude of the in-car noise in the rail corrugation section, and the regression analysis method is used to establish the mapping relationship between the main frequency amplitude of the car body vibration acceleration, the main frequency amplitude of the in-car noise and the rail corrugation amplitude to realize the identification of the rail corrugation amplitude. The mapping relationship is:

[0128] A c =ψ(A a , A p );

[0129] Among them, A c is the rail corrugation amplitude, A a A is the main frequency amplitude of the vehicle body vibration acceleration, p is the main frequency amplitude of the noise inside the car.

[0130] In this embodiment 2, for the identification of high-speed railway rail corrugation, according to the above-mentioned implementation method, the relationship between the variation of the body vibration acceleration and the in-vehicle noise with the mileage during the operation of the high-speed vehicle is obtained, the body vibration acceleration and the in-vehicle noise are divided into unit sections according to the mileage, and the length of the unit section is in the range of L / 2 to L, and the body acceleration and in-vehicle noise characteristics of each section are obtained, including: maximum value, minimum value, peak-to-peak value, kurtosis, effective value, and peak factor; a support vector machine is used to train the six time domain characteristics of the body vibration acceleration and the six time domain characteristics of the in-vehicle noise, and the grid method is used to optimize the penalty factor and the width of the radial basis of the support vector machine; the parameter penalty factor and the width of the radial basis are set in the range of 2-11 to 211, and the optimal support vector machine high-speed railway rail corrugation identification model is obtained to realize the identification of rail corrugation; and further according to the vehicle speed, the body vibration acceleration and the frequency and amplitude of the in-vehicle noise, the identification of the wavelength and amplitude of the rail corrugation is realized.

[0131] Example 3

[0132] This embodiment 3 provides a heavy-duty railway rail corrugation identification method based on a combined vibration and noise test in the driver's cab. For the identification of heavy-duty railway rail corrugation, according to the above implementation method, the relationship between the vibration acceleration of the floor of the heavy-duty railway driver's cab and the noise in the driver's cab and the mileage is obtained, the vibration acceleration and the noise in the driver's cab are divided into unit sections according to the mileage, and the length of the unit section is in the range of 5L to 10L. The acceleration characteristics of the vehicle body in each section include: maximum value, minimum value, peak-to-peak value, kurtosis, effective value, peak factor, margin factor, impulse factor and the characteristics of the vehicle interior noise include: peak-to-peak value, kurtosis, effective value, peak factor, margin factor, impulse factor; the support vector machine is used to train the 8 time domain characteristics of the vibration acceleration and the 6 time domain characteristics of the noise in the driver's cab, and the grid method is used to optimize the penalty factor and the width of the radial basis of the support vector machine; the parameter penalty factor and the width of the radial basis are set in the range of 2-9 to 29, and the optimal support vector machine heavy-duty railway rail corrugation identification model is obtained to realize the identification of heavy-duty railway rail corrugation; and further according to the vehicle speed, the vibration acceleration of the driver's cab and the frequency and amplitude of the noise in the driver's cab, the wavelength and amplitude of the heavy-duty railway rail corrugation are recognized.

[0133] Example 4

[0134] This embodiment 4 provides a method for identifying subway rail corrugation based on vehicle vibration and noise combined with multi-point testing. For the identification of subway rail corrugation, according to the above implementation, body vibration acceleration sensors and sound pressure sensors are arranged in 3 to 5 carriages of the subway vehicle, and the relationship between body vibration acceleration and noise and mileage is obtained. The body vibration acceleration and the noise inside the vehicle are divided into unit sections according to mileage, and the length of the unit section is in the range of L / 3 to L / 2. The body acceleration and the noise inside the vehicle characteristics of each section are obtained, including: maximum value, minimum value, peak-to-peak value, kurtosis, and effective value; the five time domain characteristics of the body vibration acceleration and the five time domain characteristics of the noise inside the vehicle are trained by a support vector machine, and the penalty factor and the width of the radial basis of the support vector machine are optimized by the grid method; the parameter penalty factor and the width of the radial basis are set in the range of 2-8 to 28, and the optimal support vector machine rail corrugation identification model is obtained to realize the identification of rail corrugation; and further, according to the vehicle speed, the body vibration acceleration and the frequency and amplitude of the noise inside the vehicle, the wavelength and amplitude of the rail corrugation are recognized.

[0135] Example 5

[0136] Embodiment 5 of the present invention provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium is used to store computer instructions. When the computer instructions are executed by a processor, the above-mentioned subway rail corrugation identification method based on vehicle vibration and noise combined test is implemented, and the method includes:

[0137] Acquire train operation data; wherein the operation data includes vehicle body vibration acceleration, in-vehicle noise data, operation time, speed and mileage information during operation;

[0138] Calculate the relationship between the car body vibration acceleration and the noise inside the car and the mileage during the train operation process based on the train operation data;

[0139] The vehicle body vibration acceleration and interior noise are divided into unit sections according to mileage to obtain the characteristics of vehicle body acceleration and interior noise in each section;

[0140] The pre-trained recognition model is used to process the data features of vehicle body vibration acceleration and in-vehicle noise to realize the recognition of rail corrugation; the wavelength and amplitude of rail corrugation are recognized based on the vehicle speed, vehicle body vibration acceleration and the frequency and amplitude of in-vehicle noise.

[0141] Example 6

[0142] Embodiment 6 of the present invention provides a computer program (product), including a computer program, wherein when the computer program is executed on one or more processors, the method for identifying subway rail corrugation based on vehicle vibration and noise combined test as described above is implemented, and the method includes:

[0143] Acquire train operation data; wherein the operation data includes vehicle body vibration acceleration, in-vehicle noise data, operation time, speed and mileage information during operation;

[0144] Calculate the relationship between the car body vibration acceleration and the noise inside the car and the mileage during the train operation process based on the train operation data;

[0145] The vehicle body vibration acceleration and interior noise are divided into unit sections according to mileage to obtain the characteristics of vehicle body acceleration and interior noise in each section;

[0146] The pre-trained recognition model is used to process the data features of vehicle body vibration acceleration and in-vehicle noise to realize the recognition of rail corrugation; the wavelength and amplitude of rail corrugation are recognized based on the vehicle speed, vehicle body vibration acceleration and the frequency and amplitude of in-vehicle noise.

[0147] Example 7

[0148] Embodiment 7 of the present invention provides an electronic device, comprising: a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory, so that the electronic device executes instructions for implementing the above-mentioned subway rail corrugation identification method based on vehicle vibration and noise combined test, the method comprising:

[0149] Acquire train operation data; wherein the operation data includes vehicle body vibration acceleration, in-vehicle noise data, operation time, speed and mileage information during operation;

[0150] Calculate the relationship between the car body vibration acceleration and the noise inside the car and the mileage during the train operation process based on the train operation data;

[0151] The vehicle body vibration acceleration and interior noise are divided into unit sections according to mileage to obtain the characteristics of vehicle body acceleration and interior noise in each section;

[0152] The pre-trained recognition model is used to process the data features of vehicle body vibration acceleration and in-vehicle noise to realize the recognition of rail corrugation; the wavelength and amplitude of rail corrugation are recognized based on the vehicle speed, vehicle body vibration acceleration and the frequency and amplitude of in-vehicle noise.

[0153] In summary, the subway rail corrugation identification method and system based on vehicle vibration and noise joint testing described in the embodiment of the present invention, based on vehicle speed, vehicle body vibration acceleration and in-vehicle noise characteristics, uses machine learning methods according to vehicle running speed and mileage, and can quickly, economically, conveniently and accurately identify rail corrugation and its specific characteristics, thereby providing important data support for subway operation and maintenance departments to formulate scientific and efficient maintenance strategies.

[0154] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0155] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0156] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0157] These computer program instructions can also be loaded onto a computer or other programmable data processing device, and a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0158] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative work on the basis of the technical solution disclosed in the present invention should be included in the scope of protection of the present invention.

Claims

1. A subway rail corrugation identification method based on vehicle vibration and noise combined test, characterized in that: include: Acquire train operation data; wherein the operation data includes vehicle body vibration acceleration, in-vehicle noise data, operation time, speed and mileage information during operation; Obtaining the characteristics of the vehicle acceleration and interior noise of each section includes: dividing the vehicle vibration acceleration data and the interior noise data into units, and extracting the maximum value, minimum value, peak-to-peak value, kurtosis, effective value, peak factor, margin factor, and pulse factor of the vehicle acceleration and interior noise data of each unit section; Calculate the relationship between the car body vibration acceleration and the noise inside the car and the mileage during the train operation process based on the train operation data; The vehicle body vibration acceleration and interior noise are divided into unit sections according to mileage to obtain the characteristics of vehicle body acceleration and interior noise in each section; The pre-trained recognition model is used to process the data features of the vehicle body vibration acceleration and the in-vehicle noise to realize the recognition of rail corrugation; the wavelength and amplitude of the rail corrugation are recognized according to the vehicle speed, the vehicle body vibration acceleration and the frequency and amplitude of the in-vehicle noise; the training of the recognition model includes: taking the data features of each divided unit section as a training set, selecting a radial basis kernel function as a kernel function of a support vector machine, introducing the radial basis kernel function into the support vector machine, and obtaining the optimal classification function of the support vector machine; using the grid method to optimize the penalty factor and the width of the radial basis of the support vector machine; calculating the classification accuracy, and obtaining the penalty factor and the width of the radial basis when the classification accuracy is the highest from the qualified classification accuracy as the optimal parameters, and constructing the optimal support vector machine rail corrugation recognition model; Among them, the train speed of the identified rail corrugation section is calculated, the main frequency of the car body vibration acceleration and the main frequency of the interior noise in the rail corrugation section are calculated by the Fourier transform method, and the common frequency of the main frequency of the car body vibration acceleration and the main frequency of the interior noise is taken as the rail corrugation passing frequency; the characteristic wavelength of the rail corrugation is calculated according to the train speed of the identified rail corrugation section and the rail corrugation passing frequency, so as to realize the identification of the rail corrugation wavelength; the main frequency amplitude of the car body vibration acceleration and the main frequency amplitude of the interior noise in the rail corrugation section are calculated by the Fourier transform method, and the mapping relationship between the main frequency amplitude of the car body vibration acceleration, the main frequency amplitude of the interior noise and the rail corrugation amplitude is established by the regression analysis method, so as to realize the identification of the rail corrugation amplitude.

2. The subway rail corrugation identification method based on vehicle vibration and noise combined test according to claim 1 is characterized in that: According to the changing relationship between the body vibration acceleration and the interior noise data with time, combined with the changing relationship between the train running mileage and time, time is regarded as an intermediate variable, thus obtaining the changing relationship between the body vibration acceleration and the interior noise with mileage.

3. A subway rail corrugation identification system based on vehicle vibration and noise combined testing, characterized in that: include: An acquisition module is used to acquire train operation data; The operation data includes vehicle body vibration acceleration, vehicle interior noise data, operation time, speed and mileage information during operation; Obtaining the characteristics of the vehicle acceleration and interior noise of each section includes: dividing the vehicle vibration acceleration data and the interior noise data into units, and extracting the maximum value, minimum value, peak-to-peak value, kurtosis, effective value, peak factor, margin factor, and pulse factor of the vehicle acceleration and interior noise data of each unit section; A calculation module, used to calculate the relationship between the car body vibration acceleration and the noise inside the car and the mileage during the train operation process according to the train operation data; A division module is used to divide the vehicle body vibration acceleration and the interior noise into unit sections according to mileage, and obtain the vehicle body acceleration and interior noise characteristics of each section; The recognition module is used to process the data features of the vehicle body vibration acceleration and the in-vehicle noise using a pre-trained recognition model to realize the recognition of rail corrugation; the wavelength and amplitude of the rail corrugation are recognized according to the vehicle speed, the vehicle body vibration acceleration and the frequency and amplitude of the in-vehicle noise; the training of the recognition model includes: taking the data features of each divided unit section as a training set, selecting a radial basis kernel function as a kernel function of a support vector machine, introducing the radial basis kernel function into the support vector machine, and obtaining an optimal classification function of the support vector machine; optimizing the penalty factor and the width of the radial basis of the support vector machine by a grid method; calculating the classification accuracy, and obtaining the penalty factor and the width of the radial basis when the classification accuracy is the highest from the qualified classification accuracy as the optimal parameters, and constructing an optimal support vector machine rail corrugation recognition model; Among them, the train speed of the identified rail corrugation section is calculated, the main frequency of the car body vibration acceleration and the main frequency of the interior noise in the rail corrugation section are calculated by the Fourier transform method, and the common frequency of the main frequency of the car body vibration acceleration and the main frequency of the interior noise is taken as the rail corrugation passing frequency; the characteristic wavelength of the rail corrugation is calculated according to the train speed of the identified rail corrugation section and the rail corrugation passing frequency, so as to realize the identification of the rail corrugation wavelength; the main frequency amplitude of the car body vibration acceleration and the main frequency amplitude of the interior noise in the rail corrugation section are calculated by the Fourier transform method, and the mapping relationship between the main frequency amplitude of the car body vibration acceleration, the main frequency amplitude of the interior noise and the rail corrugation amplitude is established by the regression analysis method, so as to realize the identification of the rail corrugation amplitude.

4. A computer program product, characterized in that It comprises a computer program, which, when running on one or more processors, is used to implement the subway rail corrugation identification method based on vehicle vibration and noise combined test as claimed in claim 1 or 2.

5. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium is used to store computer instructions. When the computer instructions are executed by the processor, the subway rail corrugation identification method based on vehicle vibration and noise combined testing as described in claim 1 or 2 is implemented.

6. An electronic device, characterized in that: include: A processor, a memory and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory so that the electronic device executes instructions for implementing the subway rail corrugation identification method based on vehicle vibration and noise combined testing as described in claim 1 or 2.

Images