High-speed railway wheel composite polygon wear analysis method and device
Through ACMD algorithm and envelope spectrum analysis, the frequency characteristics of wheel polygons are extracted, and the problem of confusion between wheel polygon wear detection and track wave wear in the existing technology is solved, and high-precision wheel polygon wear detection is realized, which improves the reliability of detection and real-time.
Patent Information
- Application Number
- CN202411098504.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-12
- Publication Date
- 2024-11-15
AI Technical Summary
The existing wheel polygon wear detection methods are easily confused with track wave wear, resulting in large detection errors and cannot effectively distinguish wheel polygon wear and track wave wear, affecting the safety and detection accuracy of train operation.
The ACMD algorithm is used to perform modal decomposition and time-frequency analysis of axle box acceleration data. Combined with envelope spectrum analysis, the frequency characteristics of the wheel polygon are extracted, and the file matching is confirmed through the preset wheel polygons to determine whether there are high-order polygons and whether the wheel has high-order polygons. Circular or non-round, achieving accurate detection of wheel polygon wear.
Effectively distinguish between wheel polygon wear and track wave wear, improves the accuracy of wheel polygon wear detection, reduces errors, enhances the ability to identify composite polygon features, and improves the reliability and real-time detection.
Smart Images

Figure CN118965579A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of data processing, and in particular to a method and a device for analyzing composite polygonal wear of high-speed railway wheels. Background Art
[0002] This section is intended to provide a background or context to the embodiments of the invention recited in the claims. No admission is made that the description herein is prior art by inclusion in this section.
[0003] In recent years, with the continuous development of high-speed railways, the interaction between wheels and tracks has gradually intensified, resulting in increasingly serious wheel wear, which can easily affect the reliability and stability of train operation. Wheel composite polygon refers to the composite problem of wheel out-of-roundness and wheel high-order polygons (for example, 10-sided and above), which is usually manifested as a wheel outer circle defect with uneven wear of the wheel contour along the circumferential direction. If this wheel disease cannot be detected and maintained in time, it may lead to a series of safety accidents such as train derailment and infrastructure damage.
[0004] Existing wheel polygon wear detection methods are generally divided into two modes: static and dynamic.
[0005] Traditional static testing refers to manual testing of wheels using handheld equipment. Although the testing equipment is inexpensive and has high measurement accuracy, the cost is reduced testing efficiency, increased time cost, and higher strength requirements.
[0006] Dynamic measurement can automatically complete the detection during the operation of the vehicle, which is the development direction of future railway engineering. It is divided into two modes: ground and vehicle-mounted.
[0007] Ground detection refers to installing sensors next to the track section to detect the status of the wheels by analyzing the signals sent by the monitoring instrument when the vehicle passes by. For example, the status of the polygonal wheel can be identified by analyzing the impact force obtained by the piezoelectric strain sensor installed on the rail. However, this type of detection method can only detect the wheel status when the vehicle passes through a specific track, and cannot monitor the wheel status of the entire running line in real time and track the wheel polygonization process.
[0008] Therefore, dynamic detection technology based on vehicle-mounted signals has become the main research direction now.
[0009] On-board detection refers to the technology of dynamic monitoring based on on-board detection signals of trains. For example, dynamic monitoring is performed using axle box acceleration data to analyze the potential correlation between the characteristics of axle box acceleration data and wheel polygon characteristics. However, the axle box acceleration signal corresponding to the wheel polygon has a periodic characteristic, which is similar to the track corrugation disease. Therefore, it is very easy to be confused with track corrugation in wheel polygon wear diagnosis, resulting in large errors in the current detection of wheel polygon order, profile characteristics, etc.
[0010] In summary, the existing wheel polygon wear diagnosis is easily confused with track corrugation and the wheel polygon detection error is large. Summary of the invention
[0011] The embodiment of the present invention provides a high-speed railway wheel composite polygon wear analysis method, which is used to distinguish wheel polygon wear from track corrugation and improve the high-speed railway wheel polygon wear detection accuracy. The method includes:
[0012] Collect the original axle box acceleration data of the railway;
[0013] Based on the ACMD (Adaptive Chirp Mode Decomposition) algorithm, the original axle box acceleration data is modally decomposed to obtain multiple modes;
[0014] Perform time-frequency analysis on each mode and extract the main frequency of each mode;
[0015] Matching the main frequencies of all modes with a preset wheel polygon confirmation file, and determining whether the wheel has a high-order polygon according to the matching result; the preset wheel polygon confirmation file includes a correspondence between wheel polygon values and frequencies;
[0016] Perform envelope spectrum analysis on the original axle box acceleration data to obtain the main frequency of the envelope spectrum;
[0017] Determine the calculated value of wheel circumference according to the main frequency of the envelope spectrum and the train speed;
[0018] Determine whether the wheel is round or not based on the calculated wheel circumference value and the theoretical wheel circumference value;
[0019] The wheel polygon wear result is determined based on whether the wheel has high-order polygons, and whether the wheel is round or out of round.
[0020] The embodiment of the present invention further provides a high-speed railway wheel composite polygon wear analysis device, which is used to distinguish wheel polygon wear from track corrugation and improve the wheel polygon wear detection accuracy. The device includes:
[0021] A data acquisition module is used to collect the original axle box acceleration data of the railway;
[0022] The first processing module is used to perform modal decomposition on the original axle box acceleration data based on the ACMD algorithm to obtain multiple modes; perform time-frequency analysis on each mode to extract the main frequency of each mode; match the main frequencies of all modes with a preset wheel polygon confirmation file, and determine whether the wheel has a high-order polygon based on the matching result; the preset wheel polygon confirmation file includes the corresponding relationship between the wheel polygon value and the frequency;
[0023] The second processing module is used to perform envelope spectrum analysis on the original axle box acceleration data to obtain the main frequency of the envelope spectrum; determine the calculated value of the wheel circumference according to the main frequency of the envelope spectrum and the train speed; determine whether the wheel is round or not according to the calculated value of the wheel circumference and the theoretical value of the wheel circumference;
[0024] The result output module is used to determine the wheel polygon wear result according to whether the wheel has high-order polygons and whether the wheel is round or not.
[0025] An embodiment of the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned high-speed railway wheel composite polygon wear analysis method when executing the computer program.
[0026] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned high-speed railway wheel composite polygon wear analysis method is implemented.
[0027] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the above-mentioned high-speed railway wheel composite polygon wear analysis method is implemented.
[0028] In the embodiment of the present invention, original axle box acceleration data of the railway is collected; based on the ACMD algorithm, the original axle box acceleration data is modally decomposed to obtain multiple modes; time-frequency analysis is performed on each mode to extract the main frequency of each mode; the main frequencies of all modes are matched with a preset wheel polygon confirmation file, and whether the wheel has a high-order polygon is determined according to the matching result; the preset wheel polygon confirmation file includes the corresponding relationship between the wheel polygon value and the frequency; the original axle box acceleration data is enveloped The analysis is performed to obtain the envelope spectrum main frequency; the calculated value of the wheel circumference is determined according to the envelope spectrum main frequency and the train speed; whether the wheel is round or not according to the calculated value of the wheel circumference and the theoretical value of the wheel circumference; the wheel polygon wear result is determined according to whether the wheel has a high-order polygon and whether the wheel is round or not. In the embodiment of the present invention, modal decomposition is performed using the ACMD algorithm. First, it is determined whether the wheel has a high-order polygon based on the corresponding relationship between the wheel polygon value and the frequency. Then, envelope spectrum analysis is further used to determine whether the wheel is round or not. By combining the ACMD algorithm analysis results and the envelope spectrum analysis results, the frequency characteristics of the wheel composite polygon are effectively extracted, and the wheel polygon wear is distinguished from the track corrugation. At the same time, the wheel polygon wear detection accuracy is improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:
[0030] Figure 1 It is a schematic diagram of the process of the composite polygon wear analysis method of high-speed railway wheels in an embodiment of the present invention;
[0031] Figure 2 A schematic diagram of a specific embodiment of the high-speed railway wheel composite polygon wear analysis method in an embodiment of the present invention;
[0032] Figure 3 , Figure 4 , Figure 5 They are respectively the original waveform diagram, power spectrum, and time-frequency diagram of the axle box acceleration data collected by the high-speed comprehensive detection train in an embodiment of the present invention;
[0033] Figure 6 Schematic diagram of each mode of the axle box acceleration data collected by the high-speed comprehensive detection train after ACMD processing in an embodiment of the present invention;
[0034] Figure 7 for Figure 6Time-frequency diagram of 4 modes after multiple mode screening;
[0035] Figure 8 Schematic diagram of envelope spectrum analysis of axle box acceleration data collected by a high-speed comprehensive inspection train in an embodiment of the present invention;
[0036] Fig. 9 Schematic diagram of a high-speed railway wheel composite polygonal wear analysis device in an embodiment of the present invention. DETAILED DESCRIPTION
[0037] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.
[0038] The acquisition, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws and regulations.
[0039] Although a lot of research has been done in the prior art on the formation mechanism and detection of wheel polygons, most methods are aimed at detecting wheel polygons of a single order. In addition, the axle box acceleration signal corresponding to the wheel polygon has a periodic characteristic similar to the rail corrugation disease, which is easily confused with rail corrugation in fault diagnosis. Due to the complexity and diversity of wheel problems in railway vehicles in reality, in most cases, there will be a composite wheel polygon composed of wheel out of roundness and high-order wheel polygons. There is no relevant technical description on how to extract the features of the composite wheel polygon and distinguish it from rail corrugation.
[0040] In view of the periodicity of the axle box acceleration of the wheel polygon, the embodiment of the present invention proposes a feature extraction and online rapid recognition technology for the dynamic response data of the wheel composite polygon. Figure 1 FIG. 1 is a flow chart of a method for analyzing composite polygonal wear of high-speed railway wheels according to an embodiment of the present invention. Figure 1 As shown, the method includes:
[0041] Step 101, collecting original axle box acceleration data of the railway;
[0042] Step 102: Based on the ACMD algorithm, modal decomposition is performed on the original axle box acceleration data to obtain multiple modes;
[0043] Step 103: Perform time-frequency analysis on each mode to extract the main frequency of each mode;
[0044] Step 104, matching the main frequencies of all modes with a preset wheel polygon confirmation file, and determining whether the wheel has a high-order polygon according to the matching result; the preset wheel polygon confirmation file includes a correspondence between wheel polygon values and frequencies;
[0045] Step 105, performing envelope spectrum analysis on the original axle box acceleration data to obtain the envelope spectrum main frequency;
[0046] Step 106, determining a calculated value of the wheel circumference according to the envelope spectrum main frequency and the train speed;
[0047] Step 107, determining whether the wheel is round or not according to the calculated wheel circumference value and the theoretical wheel circumference value;
[0048] Step 108: Determine the wheel polygon wear result according to whether the wheel has a high-order polygon and whether the wheel is round or not.
[0049] from Figure 1 It can be seen from the illustrated process that in the embodiment of the present invention, the original axle box acceleration data of the railway is collected; based on the ACMD algorithm, the original axle box acceleration data is modally decomposed to obtain multiple modes; time-frequency analysis is performed on each mode to extract the main frequency of each mode; the main frequencies of all modes are matched with the preset wheel polygon confirmation file, and whether the wheel has a high-order polygon is determined according to the matching result; the preset wheel polygon confirmation file includes the corresponding relationship between the wheel polygon value and the frequency; the original axle box acceleration data is subjected to envelope spectrum analysis to obtain the envelope spectrum main frequency; the calculated wheel circumference is determined according to the envelope spectrum main frequency and the train speed; whether the wheel is round or not is determined according to the calculated wheel circumference and the theoretical wheel circumference; and the wheel polygon wear result is determined according to whether the wheel has a high-order polygon and whether the wheel is round or not. In the embodiment of the present invention, modal decomposition is performed using the ACMD algorithm. First, it is determined whether the wheel has a high-order polygon based on the corresponding relationship between the wheel polygon value and the frequency. Then, envelope spectrum analysis is further used to determine whether the wheel is round or not. By combining the ACMD algorithm analysis results and the envelope spectrum analysis results, the frequency characteristics of the wheel composite polygon are effectively extracted, and the wheel polygon wear is distinguished from the track corrugation. At the same time, the wheel polygon wear detection accuracy is improved.
[0050] The polygonal wear analysis method for high-speed railway wheels in an embodiment of the present invention is explained in detail below.
[0051] First, the original axle box acceleration data of the railway is collected. During implementation, the data can be collected based on the detection equipment installed on the high-speed comprehensive detection train. The collected data is transmitted to the high-speed comprehensive detection train server in real time. The high-speed railway wheel polygonal wear analysis method in the embodiment of the present invention can be executed on the high-speed comprehensive detection train server, or the high-speed comprehensive detection train server can transmit the data to a remote server through the network, and the high-speed railway wheel polygonal wear analysis method in the embodiment of the present invention can be executed by the remote server.
[0052] For example, the original high-speed railway axle box acceleration data is x(t) = {x(t0), x(t1), …, x(t N-1 )}, where N is the number of sampling points and the sampling frequency is Fs.
[0053] Then, based on the ACMD algorithm, the original axle box acceleration data x(t)={x(0),x(t1),…,x(t N-1 )} performs modal decomposition, in which the influence of noise on the signal is eliminated.
[0054] For example, the decomposition makes ε is the noise which can be ignored, x i (t) is represented as the i-th mode, with a total of k modes.
[0055] ACMD is an efficient adaptive signal decomposition method that can estimate each signal component independently without knowing the number of signal components in advance.
[0056] A non-stationary vibration signal x(t) containing k components can be modeled as:
[0057]
[0058] Among them, A i (t) is the instantaneous amplitude, f i (τ) is the instantaneous frequency, is the initial phase of the ith mode.
[0059] After demodulation technology, x(t) can be converted into:
[0060]
[0061] Among them, α i (t), β i (t) are two demodulation operators, for and The corresponding frequency function is:
[0062]
[0063] Among them, θ i For phase.
[0064] The ACMD algorithm can be transformed into solving the i-th mode of the original signal (original axle box acceleration data):
[0065]
[0066] Among them, ε is the weight coefficient.
[0067] Since the original data of the high-speed railway axle box acceleration is a discrete signal x(t) = {x(t0), x(t1), …, x(t N-1 )}, so the above formula can be transformed into solving The minimum value of , where:
[0068] x=[x(t0),…,x(t N-1 )] T ;
[0069]
[0070] D is a second-order difference matrix of size (N-2)×N;
[0071]
[0072] α k =[α k (t0),…,α k (t N-1 )] T
[0073] β k =[β k (t0),…,β k (t N-1 )] T
[0074] C k =diag[cos(θ k (t0)),…,cos(θ k (t N-1 ))]
[0075] S k =diag[sin(θ k (t0)),…,sin(θ k (t N-1 ))]
[0076]
[0077] The ACMD decomposition method for the original axle box acceleration data of high-speed railway is described as follows:
[0078] (1) Update the combined delinear frequency modulation component z through an iterative process k :
[0079]
[0080] where i corresponds to the number of iterations. Therefore, the desired modal sequence can be rewritten as:
[0081]
[0082] (2) According to the arc tangent demodulation, we can get the kth IF f k The increment is:
[0083]
[0084] (3) Generally speaking, IF should be a low-pass function. In order to reduce the impact of noise, the increment of instantaneous frequency should meet the low-pass condition:
[0085]
[0086] in μ>0 is the weight parameter, is an estimate of the instantaneous frequency increment:
[0087]
[0088] Where I is the identity matrix. Then the kth instantaneous frequency f is updated through an iterative process k :
[0089]
[0090] (4) Obtained through step 3 Calculating the Matrix
[0091] (5) Repeat steps 1-4 until the difference between two consecutive iterations is negligible, and then the target mode is obtained.
[0092] (6) Extract each required mode by repeating steps 1-5. The stopping condition is satisfied: where r k is the residual component, and δ is the set threshold.
[0093] In one embodiment, each mode has a signal energy property value;
[0094] Before performing time-frequency analysis on each mode and extracting the main frequency of each mode, the following can also be included:
[0095] The signal energy attribute values are sorted in descending order. If the sum of the signal energy attribute values of the current first numerical value modes is greater than the preset threshold, the first numerical value mode is screened out from multiple modes for subsequent processing; wherein the preset threshold is, for example, 90%, 95%, etc., which can be set according to actual conditions.
[0096] For example, sort all modes by signal energy attribute value, and select the smallest positive integer k such that Where E is the sum of the energy attribute values of each modal signal.
[0097] When implemented, each mode x is calculated as follows i (t) signal energy attribute value E i :
[0098]
[0099] Get the energy distribution Choose the smallest positive integer k such that
[0100] Then, for each mode x i (t) Perform time-frequency analysis.
[0101] In one embodiment, performing time-frequency analysis on each mode to extract the main frequency of each mode may include:
[0102] Perform time-frequency analysis on each mode to obtain the time-frequency diagram of each mode;
[0103] The frequency of the light band concentrated area in the time-frequency diagram of each mode is extracted as the main frequency of each mode.
[0104] In step 104, the main frequencies of all modes are matched with a preset wheel polygon confirmation file, and whether the wheel has a high-order polygon is determined based on the matching result; the preset wheel polygon confirmation file includes the correspondence between the wheel polygon value and the frequency.
[0105] During implementation, the frequency is converted into wavelength to determine whether the wheel has a high-order wheel polygon.
[0106] In one embodiment, the correspondence between the wheel polygon value and the frequency can be determined as follows:
[0107] Given a wheel polygon value x, determine the frequency y corresponding to the wheel polygon value x according to the following relationship:
[0108] Theoretical wheel circumference / wavelength = x
[0109] Wavelength = train speed / y
[0110] In the formula, wavelength is an intermediate variable used for calculation.
[0111] In the embodiment of the present invention, after the main frequencies of the multiple modes are obtained, each main frequency is compared with the frequency corresponding to the wheel polygon value, so as to determine whether a high-order polygon exists.
[0112] In one embodiment, matching the main frequencies of all modes with a preset wheel polygon confirmation file, and determining whether the wheel has a high-order polygon according to the matching result may include:
[0113] Eliminate the natural frequency of the train tooling and the modulated frequency from the main frequencies of all modes to obtain the main frequency of the screened mode;
[0114] The main frequency of the screened mode is matched with the preset wheel polygon confirmation file. When the main frequency of the screened mode and the frequency in the preset wheel polygon confirmation file are less than the preset frequency threshold, it is determined that the wheel has a high-order polygon.
[0115] For example, if there are multiple modal main frequencies, the main frequencies that are obviously not the corresponding polygons (such as the main frequencies that are obviously modulated, the natural frequencies of the tooling, etc.) can be eliminated. When there is a main frequency that is obviously close to or equal to the frequency corresponding to the wheel polygon value, it is determined that the wheel has a high-order polygon that responds to the wheel polygon value.
[0116] Based on the power spectrum and time-frequency diagram of the original high-speed railway axle box acceleration data, it can be found that the frequency peaks are mainly above 200Hz. The low-frequency characteristics corresponding to the wheel out-of-round phenomenon cannot be directly observed due to the influence of the modulation phenomenon. Therefore, in order to further improve the accuracy of wheel composite polygon detection, the embodiment of the present invention introduces envelope spectrum analysis to target frequency characteristics that are not easy to find in the original waveform.
[0117] The envelope spectrum of a signal is an analytical method that describes the energy distribution of the signal envelope in the frequency domain. The envelope spectrum provides the modulation frequency information carried in the signal envelope. In a modulated signal, the envelope spectrum reveals the changes in the modulation frequency and reflects the dynamic characteristics of the information carried by the signal.
[0118] In the embodiment of the present invention, envelope spectrum analysis is performed on the original signal to obtain the main frequency of the envelope spectrum; a calculated value of the wheel circumference is determined according to the main frequency of the envelope spectrum and the train speed; and whether the wheel is round or not is determined according to the calculated value of the wheel circumference and the theoretical value of the wheel circumference.
[0119] In one embodiment, performing envelope spectrum analysis on the original axle box acceleration data to obtain the envelope spectrum main frequency may include:
[0120] Perform Hilbert transform on the original axle box acceleration data to obtain multiple sequences;
[0121] Calculate the instantaneous amplitude of all sequences to obtain multiple envelope signals;
[0122] Performing discrete Fourier transform on multiple envelope signals to obtain a spectrum estimation value of each envelope signal;
[0123] The main frequency of the envelope spectrum is determined according to the frequency spectrum estimation values of all envelope signals.
[0124] When implementing it specifically, it can be achieved by following the steps below.
[0125] (1) The original axle box acceleration data of the high-speed railway is x(t) = {x(t0), x(t1), …, x(t N-1 )};
[0126] (2) The original signal x(t) is Hilbert transformed to obtain the sequence h(t). The calculation formula of h(t) is:
[0127]
[0128] (3) Obtaining the envelope signal z[t]=|x[t]+j·h[t]|;
[0129] (4) The envelope signal z[t] is divided into equally spaced segments z0[t], z1[t], ..., z n [t];
[0130] (5) Perform discrete Fourier transform on each segment to obtain the spectrum estimate P of each segment XX,i (f), expressed as: Where X i (f) is z i The discrete Fourier transform of [n];
[0131] (6) The spectrum estimates of all segments are averaged to obtain the final power spectrum estimate.
[0132]
[0133] Where M is the total number of segments.
[0134] In one embodiment, determining whether a wheel is round or not according to the calculated wheel circumference value and the theoretical wheel circumference value may include:
[0135] Calculate the difference between the calculated wheel circumference value and the theoretical wheel circumference value;
[0136] Based on the difference and the preset error value, determine whether the wheel is round or out of round.
[0137] The preset error value is determined according to actual conditions, for example, 0.01m.
[0138] Figure 2 FIG. 1 is a schematic diagram of a specific embodiment of the high-speed railway wheel composite polygon wear analysis method according to an embodiment of the present invention, with reference to FIG. Figure 2 In the embodiment of the present invention, the axle box acceleration data is divided into two parts for processing, one part is ACMD algorithm processing, and the other part is envelope spectrum analysis. Finally, the processing results of the two parts are combined to determine the wheel composite polygon result, one part of which is subjected to ACMD modal decomposition and energy screening to obtain multiple target modes, and then video analysis is performed in turn. According to the correlation between frequency, wavelength, and theoretical value of wheel circumference, it is determined whether there is a high-order polygon or a high-order polygon value, and the other part is subjected to envelope spectrum analysis to determine whether the wheel is round or not.
[0139] The following is a data analysis using a specific embodiment, and the processing is explained using acceleration signals collected by a high-speed comprehensive inspection train on a high-speed railway line.
[0140] Figure 3 , Figure 4 , Figure 5 They are respectively the original waveform diagram, power spectrum, and time-frequency diagram of the axle box acceleration data collected by the high-speed comprehensive inspection train in an embodiment of the present invention. It can be seen that the axle box acceleration data cannot intuitively show the characteristics of the wheel polygon and is easily confused with corrugation.
[0141] The method of the embodiment of the present invention is used to use the ACMD algorithm to Figures 3 to 5 The original signal is decomposed, and the result is as follows Figure 6 As shown, Figure 6 The following is a schematic diagram of each mode after ACMD processing of the axle box acceleration data collected by the high-speed comprehensive detection train in the embodiment of the present invention. According to calculations, the sum of the energy of the first four modes is greater than 90% of the total energy, so time-frequency analysis is performed on the first four modes. The time-frequency diagram is as follows: Figure 7 As shown, Figure 7 for Figure 6 The time-frequency diagram of the four modes after multiple mode screening. It is found that the light band of the first mode is mainly concentrated at 1333.01Hz, the light band of the second mode is mainly concentrated at 1833.50Hz, the light band of the third mode is mainly concentrated at 1583.25Hz, and the light band of the fourth mode is mainly concentrated at 250.24Hz. Compared with the time-frequency diagram of the original waveform, the time-frequency diagram of each mode after ACMD can better show the time-frequency characteristics and modulation phenomenon of the signal. Among them, the first and second modes are obviously modulated, and the third mode is the natural frequency of the tooling. Therefore, the calculation is based on the fourth mode. The corresponding wavelength of the fourth mode is calculated to be 0.26m, which is 1 / 11 of the circumference of the wheel. It is judged that the wheel has an 11th-order polygon.
[0142] Then, envelope spectrum analysis is used to assist in extracting the wheel composite polygon features. Figure 8Schematic diagram of envelope spectrum analysis of axle box acceleration data collected by a high-speed comprehensive inspection train in an embodiment of the present invention, Figure 8 The envelope spectrum of the original axle box acceleration data is shown. Figure 8 It can be seen that the main frequency of the envelope spectrum in this section is 22.61Hz, and the corresponding wavelength is 2.85m, which is equal to the circumference of the wheel, that is, the wheel is not round. Here, for example, the main frequency is converted into wavelength and represented by a, and the circumference of the wheel is represented by b. When b / a is a positive integer (a small error may exist), it means that the wheel is not round, otherwise, it means that the wheel is round.
[0143] Therefore, combining the above two processing results, the wheel in this example is not round and has an 11th-order polygon.
[0144] In summary, although both the wheel composite polygonal features and corrugation are periodic, the embodiments of the present invention can effectively extract the frequency features of the wheel composite polygonal features and distinguish them from corrugation through the ACMD algorithm and envelope spectrum analysis.
[0145] The present invention also provides a high-speed railway wheel composite polygon wear analysis device, as described in the following embodiments. Since the principle of solving the problem by the device is similar to the high-speed railway wheel composite polygon wear analysis method, the implementation of the device can refer to the implementation of the high-speed railway wheel composite polygon wear analysis method, and the repeated parts will not be repeated.
[0146] Fig. 9 FIG. 1 is a schematic diagram of a high-speed railway wheel composite polygon wear analysis device according to an embodiment of the present invention. Fig. 9 As shown, the device comprises:
[0147] The data acquisition module 901 is used to collect the original axle box acceleration data of the railway;
[0148] The first processing module 902 is used to perform modal decomposition on the original axle box acceleration data based on the ACMD algorithm to obtain multiple modes; perform time-frequency analysis on each mode to extract the main frequency of each mode; match the main frequencies of all modes with a preset wheel polygon confirmation file, and determine whether the wheel has a high-order polygon according to the matching result; the preset wheel polygon confirmation file includes the corresponding relationship between the wheel polygon value and the frequency;
[0149] The second processing module 903 is used to perform envelope spectrum analysis on the original axle box acceleration data to obtain the main frequency of the envelope spectrum; determine the calculated value of the wheel circumference according to the main frequency of the envelope spectrum and the train speed; determine whether the wheel is round or not according to the calculated value of the wheel circumference and the theoretical value of the wheel circumference;
[0150] The result output module 904 is used to determine the wheel polygon wear result according to whether the wheel has a high-order polygon and whether the wheel is round or not.
[0151] In one embodiment, each mode has a signal energy property value;
[0152] Also includes:
[0153] The modal screening module is used to perform time-frequency analysis on each mode in the first processing module, and sort the signal energy attribute values from large to small before extracting the main frequency of each mode. If the sum of the signal energy attribute values of the current first numerical value of modes is greater than a preset threshold, the first numerical value of modes will be screened out from multiple modes for subsequent processing.
[0154] In one embodiment, the first processing module is specifically used for:
[0155] Perform time-frequency analysis on each mode to obtain the time-frequency diagram of each mode;
[0156] The frequency of the light band concentrated area in the time-frequency diagram of each mode is extracted as the main frequency of each mode.
[0157] In one embodiment, the correspondence between the wheel polygon value and the frequency is determined as follows:
[0158] Given a wheel polygon value x, determine the frequency y corresponding to the wheel polygon value x according to the following relationship:
[0159] Theoretical wheel circumference / wavelength = x
[0160] Wavelength = train speed / y
[0161] In the formula, wavelength is an intermediate variable used for calculation.
[0162] In one embodiment, the first processing module is specifically used for:
[0163] Eliminate the natural frequency of the train tooling and the modulated frequency from the main frequencies of all modes to obtain the main frequency of the screened mode;
[0164] The main frequency of the screened mode is matched with the preset wheel polygon confirmation file. When the difference between the main frequency of the screened mode and the frequency in the preset wheel polygon confirmation file is less than the preset frequency threshold, it is determined that the wheel has a high-order polygon.
[0165] In one embodiment, the second processing module is specifically configured to:
[0166] Perform Hilbert transform on the original axle box acceleration data to obtain multiple sequences;
[0167] Calculate the instantaneous amplitude of all sequences to obtain multiple envelope signals;
[0168] Performing discrete Fourier transform on multiple envelope signals to obtain a spectrum estimation value of each envelope signal;
[0169] The main frequency of the envelope spectrum is determined according to the frequency spectrum estimation values of all envelope signals.
[0170] In one embodiment, the second processing module is specifically configured to:
[0171] Calculate the difference between the calculated wheel circumference value and the theoretical wheel circumference value;
[0172] Based on the difference and the preset error value, determine whether the wheel is round or out of round.
[0173] An embodiment of the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned high-speed railway wheel composite polygon wear analysis method when executing the computer program.
[0174] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned high-speed railway wheel composite polygon wear analysis method is implemented.
[0175] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the above-mentioned high-speed railway wheel composite polygon wear analysis method is implemented.
[0176] In the embodiment of the present invention, original axle box acceleration data of the railway is collected; based on the ACMD algorithm, the original axle box acceleration data is modally decomposed to obtain multiple modes; time-frequency analysis is performed on each mode to extract the main frequency of each mode; the main frequencies of all modes are matched with a preset wheel polygon confirmation file, and whether the wheel has a high-order polygon is determined according to the matching result; the preset wheel polygon confirmation file includes the corresponding relationship between the wheel polygon value and the frequency; the original axle box acceleration data is enveloped The analysis is performed to obtain the envelope spectrum main frequency; the calculated value of the wheel circumference is determined according to the envelope spectrum main frequency and the train speed; whether the wheel is round or not according to the calculated value of the wheel circumference and the theoretical value of the wheel circumference; the wheel polygon wear result is determined according to whether the wheel has a high-order polygon and whether the wheel is round or not. In the embodiment of the present invention, modal decomposition is performed using the ACMD algorithm. First, it is determined whether the wheel has a high-order polygon based on the corresponding relationship between the wheel polygon value and the frequency. Then, envelope spectrum analysis is further used to determine whether the wheel is round or not. By combining the ACMD algorithm analysis results and the envelope spectrum analysis results, the frequency characteristics of the wheel composite polygon are effectively extracted, and the wheel polygon wear is distinguished from the track corrugation. At the same time, the wheel polygon wear detection accuracy is improved.
[0177] 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.
[0178] 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.
[0179] 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 comprising 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.
[0180] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions 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.
[0181] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.
Claims
1. A high-speed railway wheel composite polygon wear analysis method, characterized in that: include: Collect the original axle box acceleration data of the railway; Based on the adaptive linear frequency modulation mode decomposition (ACMD) algorithm, the original axle box acceleration data is modally decomposed to obtain multiple modes. Perform time-frequency analysis on each mode and extract the main frequency of each mode; Matching the main frequencies of all modes with a preset wheel polygon confirmation file, and determining whether the wheel has a high-order polygon according to the matching result; the preset wheel polygon confirmation file includes a correspondence between wheel polygon values and frequencies; Perform envelope spectrum analysis on the original axle box acceleration data to obtain the main frequency of the envelope spectrum; Determine the calculated value of wheel circumference according to the main frequency of the envelope spectrum and the train speed; Determine whether the wheel is round or not based on the calculated wheel circumference value and the theoretical wheel circumference value; The wheel polygon wear result is determined based on whether the wheel has high-order polygons, and whether the wheel is round or out of round.
2. The method according to claim 1, characterized in that Each mode has a signal energy attribute value; Perform time-frequency analysis on each mode and extract the main frequency of each mode, including: The signal energy attribute values are sorted in descending order. If the sum of the signal energy attribute values of the current first numerical value modes is greater than a preset threshold, the first numerical value modes are screened out from multiple modes for subsequent processing.
3. The method according to claim 1, characterized in that Perform time-frequency analysis on each mode and extract the main frequency of each mode, including: Perform time-frequency analysis on each mode to obtain the time-frequency diagram of each mode; The frequency of the light band concentrated area in the time-frequency diagram of each mode is extracted as the main frequency of each mode.
4. The method according to claim 1, characterized in that The corresponding relationship between the wheel polygon value and the frequency is determined as follows: Given a wheel polygon value x, determine the frequency y corresponding to the wheel polygon value x according to the following relationship: Theoretical wheel circumference / wavelength = x Wavelength = train speed / y In the formula, wavelength is an intermediate variable used for calculation.
5. The method according to claim 4, characterized in that Match the main frequencies of all modes with the preset wheel polygon confirmation file, and determine whether the wheel has high-order polygons based on the matching results, including: Eliminate the natural frequency of the train tooling and the modulated frequency from the main frequencies of all modes to obtain the main frequency of the screened mode; The main frequency of the screened mode is matched with the preset wheel polygon confirmation file. When the difference between the main frequency of the screened mode and the frequency in the preset wheel polygon confirmation file is less than the preset frequency threshold, it is determined that the wheel has a high-order polygon.
6. The method according to claim 1, characterized in that Perform envelope spectrum analysis on the original axle box acceleration data to obtain the main frequency of the envelope spectrum, including: Perform Hilbert transform on the original axle box acceleration data to obtain multiple sequences; Calculate the instantaneous amplitude of all sequences to obtain multiple envelope signals; Performing discrete Fourier transform on multiple envelope signals to obtain a spectrum estimation value of each envelope signal; The main frequency of the envelope spectrum is determined according to the frequency spectrum estimation values of all envelope signals.
7. The method according to claim 1, characterized in that Determine whether the wheel is round or not based on the calculated wheel circumference and the theoretical wheel circumference, including: Calculate the difference between the calculated wheel circumference value and the theoretical wheel circumference value; Based on the difference and the preset error value, determine whether the wheel is round or out of round.
8. A high-speed railway wheel composite polygon wear analysis device, characterized in that: include: A data acquisition module is used to collect the original axle box acceleration data of the railway; The first processing module is used to perform modal decomposition on the original axle box acceleration data based on the ACMD algorithm to obtain multiple modes; perform time-frequency analysis on each mode to extract the main frequency of each mode; match the main frequencies of all modes with a preset wheel polygon confirmation file, and determine whether the wheel has a high-order polygon based on the matching result; the preset wheel polygon confirmation file includes the corresponding relationship between the wheel polygon value and the frequency; The second processing module is used to perform envelope spectrum analysis on the original axle box acceleration data to obtain the main frequency of the envelope spectrum; and determine the calculated value of the wheel circumference according to the main frequency of the envelope spectrum and the train speed; Determine whether the wheel is round or not based on the calculated wheel circumference value and the theoretical wheel circumference value; The result output module is used to determine the wheel polygon wear result according to whether the wheel has high-order polygons and whether the wheel is round or not.
9. The device according to claim 8, characterized in that Each mode has a signal energy attribute value; Also includes: The modal screening module is used to perform time-frequency analysis on each mode in the first processing module, and sort the signal energy attribute values from large to small before extracting the main frequency of each mode. If the sum of the signal energy attribute values of the current first numerical value of modes is greater than a preset threshold, the first numerical value of modes will be screened out from multiple modes for subsequent processing.
10. The device according to claim 8, characterized in that The first processing module is specifically used for: Perform time-frequency analysis on each mode to obtain the time-frequency diagram of each mode; The frequency of the light band concentrated area in the time-frequency diagram of each mode is extracted as the main frequency of each mode.
11. The device according to claim 8, characterized in that The corresponding relationship between the wheel polygon value and the frequency is determined as follows: Given a wheel polygon value x, determine the frequency y corresponding to the wheel polygon value x according to the following relationship: Theoretical wheel circumference / wavelength = x Wavelength = train speed / y In the formula, wavelength is an intermediate variable used for calculation.
12. The device according to claim 11, characterized in that The first processing module is specifically used for: Eliminate the natural frequency of the train tooling and the modulated frequency from the main frequencies of all modes to obtain the main frequency of the screened mode; The main frequency of the screened mode is matched with the preset wheel polygon confirmation file. When the difference between the main frequency of the screened mode and the frequency in the preset wheel polygon confirmation file is less than the preset frequency threshold, it is determined that the wheel has a high-order polygon.
13. The device according to claim 8, characterized in that The second processing module is specifically used for: Perform Hilbert transform on the original axle box acceleration data to obtain multiple sequences; Calculate the instantaneous amplitude of all sequences to obtain multiple envelope signals; Performing discrete Fourier transform on multiple envelope signals to obtain a spectrum estimation value of each envelope signal; The main frequency of the envelope spectrum is determined according to the frequency spectrum estimation values of all envelope signals.
14. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.
15. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.
16. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.
