Methods and devices for identifying and correcting local anomalies in the lateral acceleration of train bodies in the turnout area

Abstract

This invention discloses a method and apparatus for identifying and correcting local anomalies in the lateral acceleration data of a car body in a turnout area. The method includes: calculating the moving standard deviation of the lateral acceleration sequence data of the car body in the turnout area to be processed; comparing the moving standard deviation with a first threshold; when the moving standard deviation exceeds the first threshold, calculating the number of valid zero points; wherein the lateral acceleration sequence data of the car body in the turnout area to be processed is divided into multiple intervals, the lateral acceleration amplitude at the two endpoints of any interval is 0, and the number of valid zero points is the total number of endpoints of the interval whose maximum lateral acceleration amplitude exceeds a second threshold; when the number of valid zero points is lower than a third threshold, it is determined that there is high-frequency interference in the lateral acceleration sequence data of the car body in the turnout area to be processed; performing high-frequency interference filtering and signal reconstruction to obtain the corrected lateral acceleration of the car body. This invention can improve the accuracy of identifying abnormal lateral acceleration data of the car body.

Description

Technical Field

[0001] This invention relates to the field of high-speed railway engineering, and in particular to a method and device for identifying and correcting local anomalies in the lateral acceleration of the car body in the turnout area. Background Technology

[0002] This section is intended to provide background or context for the embodiments of the invention set forth in the claims. The description herein is not an admission that it is prior art simply because it is included in this section.

[0003] As a crucial component of high-speed railways, the track system provides support and guidance for operating trains, and its condition directly impacts train operation safety and passenger comfort. Therefore, monitoring the track system's condition is of paramount importance. Currently, track system condition monitoring methods are mainly divided into manual static monitoring and dynamic monitoring using inspection vehicles. Manual static monitoring is time-consuming and inefficient, relying heavily on the experience of on-site workers; dynamic monitoring, on the other hand, is highly efficient, has strong coverage, and provides more comprehensive and objective information from the data. With the continuous development of big data technology, scientifically mining the information contained in massive amounts of monitoring data can help railway professionals gain a clearer and deeper understanding of the vibration mechanisms of highly coupled and highly nonlinear railway vehicle-track coupled systems.

[0004] Currently, track condition is generally evaluated from two perspectives: track geometry and vehicle vibration. Vehicle vibration evaluation indicators include two aspects: vertical vibration acceleration and lateral vibration acceleration. Lateral vibration acceleration describes the lateral vibration state of the train during operation, and is currently assessed by evaluating its half-peak amplitude to assist in track condition evaluation. Besides peak value information, lateral vibration acceleration can also reflect train motion information, including lateral vibration intensity and throughput performance. In practical analysis, it has been found that sensors may malfunction due to external factors or their own inherent conditions. Lateral acceleration data acquired under abnormal operating conditions exhibit significant anomalies, such as baseline shift and the presence of high-frequency interference components. When evaluating the half-peak value, these anomalies have a relatively small impact on the evaluation results. However, when using other methods such as energy, entropy, and power spectrum to quantitatively describe the data characteristics of the signal, a small portion of the results will be significantly distorted. This distorted data will affect the statistical results and reduce the reliability of the lateral vibration law of the EMU.

[0005] Existing methods for identifying abnormal lateral acceleration data of railway vehicles generally employ simple filters to remove obvious outliers. However, these methods have low accuracy in identifying abnormal lateral acceleration data, resulting in poor reliability of the analysis results.

[0006] In summary, accurately identifying abnormal lateral acceleration data of the vehicle body and correcting it using appropriate methods is of great significance for the scientific analysis of lateral vibration acceleration data of the vehicle body and the understanding of the lateral vibration law of EMU trains. Therefore, this invention proposes a method for identifying and correcting abnormal lateral acceleration data of the vehicle body containing high-frequency interference components. Summary of the Invention

[0007] The technical problem to be solved by the embodiments of the present invention is to identify abnormal data of lateral acceleration of railway vehicle bodies in turnout areas containing high-frequency interference components and to make scientific and effective corrections, so as to ensure the validity and reliability of the results obtained when analyzing the lateral acceleration data of the vehicle body.

[0008] Specific issues include:

[0009] (1) A method for identifying abnormal vehicle body lateral acceleration containing high-frequency interference components;

[0010] (2) Method for correcting the lateral acceleration of abnormal vehicle bodies containing high-frequency interference components.

[0011] This invention provides a method for identifying and correcting local anomalies in the lateral acceleration data of a car body in a turnout area, in order to improve the accuracy of identifying abnormal lateral acceleration data of the car body. The method includes:

[0012] Calculate the moving standard deviation of the lateral acceleration sequence data of the car body in the turnout area to be processed;

[0013] Compare the moving standard deviation with the first threshold;

[0014] When the moving standard deviation result exceeds the first threshold, the number of valid zero points is calculated; wherein, the lateral acceleration sequence data of the turnout area to be processed is divided into multiple intervals, the lateral acceleration amplitude at the two endpoints of any interval is 0, and the number of valid zero points is the total number of the intervals whose maximum lateral acceleration amplitude exceeds the second threshold before the first endpoint.

[0015] When the number of valid zero points is lower than the third threshold, it is determined that there is high-frequency interference in the lateral acceleration sequence data of the turnout area to be processed.

[0016] The high-frequency interference filtering and signal reconstruction are performed on the lateral acceleration sequence data of the car body in the turnout area to obtain the corrected lateral acceleration of the car body.

[0017] This invention also provides a device for identifying and correcting local anomalies in the lateral acceleration of a car body in a turnout area, to improve the accuracy of identifying abnormal lateral acceleration data of the car body. The device includes:

[0018] An anomaly detection module is used to calculate the moving standard deviation of the lateral acceleration sequence data of the turnout area to be processed; compare the moving standard deviation with a first threshold; when the moving standard deviation exceeds the first threshold, calculate the number of valid zero points; wherein, the lateral acceleration sequence data of the turnout area to be processed is divided into multiple intervals, and the lateral acceleration amplitude at the two endpoints of any interval is 0, and the number of valid zero points is the total number of intervals whose maximum lateral acceleration amplitude exceeds the second threshold before the first endpoint; when the number of valid zero points is lower than a third threshold, it is determined that the lateral acceleration sequence data of the turnout area to be processed has high-frequency interference;

[0019] The correction processing module is used to perform high-frequency interference filtering and signal reconstruction on the lateral acceleration sequence data of the car body in the turnout area to obtain the corrected lateral acceleration of the car body.

[0020] This invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-mentioned method for identifying and correcting local abnormal data of lateral acceleration of the vehicle body in the turnout area.

[0021] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for processing lateral acceleration data of railway vehicle bodies.

[0022] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-mentioned method for identifying and correcting local abnormal data of lateral acceleration of the vehicle body in the turnout area.

[0023] In this embodiment of the invention, the moving standard deviation of the lateral acceleration sequence data of the car body in the turnout area to be processed is calculated; the moving standard deviation is compared with a first threshold; when the moving standard deviation exceeds the first threshold, the number of valid zeros is calculated; wherein, the lateral acceleration sequence data of the car body in the turnout area to be processed is divided into multiple intervals, and the lateral acceleration amplitude at the two endpoints of any interval is 0, and the number of valid zeros is the total number of the intervals whose maximum lateral acceleration amplitude exceeds the second threshold before the first endpoint; when the number of valid zeros is lower than a third threshold, it is determined that there is high-frequency interference in the lateral acceleration sequence data of the car body in the turnout area to be processed; high-frequency interference filtering and signal reconstruction are performed on the lateral acceleration sequence data of the car body in the turnout area to be processed to obtain the corrected lateral acceleration of the car body. In this embodiment of the invention, the moving standard deviation of the lateral acceleration sequence data of the car body in the turnout area to be processed is first calculated to determine whether the lateral acceleration sequence data of the car body in the turnout area to be processed is abnormal data and has the possibility of high-frequency interference. When the moving standard deviation result exceeds the first threshold, it indicates that the lateral acceleration sequence data of the car body in the turnout area to be processed is abnormal data and has a high possibility of high-frequency interference. The effective number of zero points is further calculated. When the effective number of zero points is lower than the third threshold, it indicates that the lateral acceleration sequence data of the car body in the turnout area to be processed has high-frequency interference. This method effectively improves the accuracy of identifying abnormal lateral acceleration data of the car body, thereby correcting the lateral acceleration sequence data of the car body in the turnout area to be processed, and ensuring the validity and reliability of the results obtained when analyzing the lateral acceleration data of the car body. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0025] Figure 1 This is a flowchart illustrating the method for identifying and correcting local anomalies in the lateral acceleration of the car body in the turnout area according to an embodiment of the present invention.

[0026] Figure 2 , Figure 3 These are, respectively, abnormal vehicle body lateral acceleration data containing high-frequency interference components and normal vehicle body lateral acceleration data in the embodiments of the present invention;

[0027] Figure 4 This is a magnified view of abnormal lateral acceleration data of the vehicle body containing high-frequency interference components in an embodiment of the present invention;

[0028] Figure 5 This is the original waveform of the lateral acceleration of the car body when the turnout passes through in an embodiment of the present invention;

[0029] Figure 6 This is a specific embodiment of the method for identifying and correcting local anomalies in the lateral acceleration of the car body in the turnout area according to the present invention;

[0030] Figure 7 This is a verification illustration of the method for identifying and correcting local anomalies in the lateral acceleration of the car body in the turnout area, as described in this embodiment of the invention. Figure 1 ;

[0031] Figure 8 This is a verification illustration of the method for identifying and correcting local anomalies in the lateral acceleration of the car body in the turnout area, as described in this embodiment of the invention. Figure 1 ;

[0032] Figure 9 This is a schematic diagram of the device for identifying and correcting local abnormal data of lateral acceleration of the vehicle body in the turnout area in an embodiment of the present invention. Detailed Implementation

[0033] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0034] Currently, high-speed integrated inspection trains are equipped with track geometry detection systems to detect the geometric shape and position of the track, as well as the vertical and lateral vibration acceleration of the train body when it passes through the current track section. This paper uses the lateral acceleration of the train body collected by the high-speed integrated inspection train as an example to illustrate the specific implementation of this invention.

[0035] Figure 1 This is a flowchart illustrating the method for identifying and correcting local anomalies in the lateral acceleration of the car body in the turnout area, as described in this embodiment of the invention. Figure 1 As shown, the method includes:

[0036] Step 101: Calculate the moving standard deviation of the lateral acceleration sequence data of the car body in the turnout area to be processed;

[0037] Step 102: Compare the moving standard deviation with the first threshold;

[0038] Step 103: When the moving standard deviation result value exceeds the first threshold, calculate the number of valid zero points; wherein, the lateral acceleration sequence data of the turnout area to be processed is divided into multiple intervals, the lateral acceleration amplitude at the two endpoints of any interval is 0, and the number of valid zero points is the total number of the intervals whose maximum lateral acceleration amplitude exceeds the second threshold before the first endpoint.

[0039] Step 104: When the number of valid zero points is lower than the third threshold, it is determined that there is high-frequency interference in the lateral acceleration sequence data of the turnout area to be processed;

[0040] Step 105: Perform high-frequency interference filtering and signal reconstruction on the lateral acceleration sequence data of the car body in the turnout area to obtain the corrected lateral acceleration of the car body.

[0041] from Figure 1 As shown in the flowchart, in this embodiment of the invention, the moving standard deviation of the lateral acceleration sequence data of the turnout area to be processed is calculated; the moving standard deviation is compared with a first threshold; when the moving standard deviation exceeds the first threshold, the number of valid zeros is calculated; wherein, the lateral acceleration sequence data of the turnout area to be processed is divided into multiple intervals, and the lateral acceleration amplitude at the two endpoints of any interval is 0, and the number of valid zeros is the total number of the intervals whose maximum lateral acceleration amplitude exceeds the second threshold before the first endpoint; when the number of valid zeros is lower than the third threshold, it is determined that there is high-frequency interference in the lateral acceleration sequence data of the turnout area to be processed; high-frequency interference filtering and signal reconstruction are performed on the lateral acceleration sequence data of the turnout area to be processed to obtain the corrected lateral acceleration of the car body. In this embodiment of the invention, the moving standard deviation of the lateral acceleration sequence data of the car body in the turnout area to be processed is first calculated to determine whether the lateral acceleration sequence data of the car body in the turnout area to be processed is abnormal data and has the possibility of high-frequency interference. When the moving standard deviation result exceeds the first threshold, it indicates that the lateral acceleration sequence data of the car body in the turnout area to be processed is abnormal data and has a high possibility of high-frequency interference. The effective number of zero points is further calculated. When the effective number of zero points is lower than the third threshold, it indicates that the lateral acceleration sequence data of the car body in the turnout area to be processed has high-frequency interference. This method effectively improves the accuracy of identifying abnormal lateral acceleration data of the car body, thereby correcting the lateral acceleration sequence data of the car body in the turnout area to be processed, and ensuring the validity and reliability of the results obtained when analyzing the lateral acceleration data of the car body.

[0042] In this embodiment of the invention, the abnormal data is caused by sensor malfunction, which manifests as high-frequency interference components in the lateral acceleration data of railway vehicles.

[0043] This invention identifies anomalous lateral acceleration data containing high-frequency interference components in the lateral acceleration data of the turnout area by calculating the standard deviation of the lateral acceleration movement and the number of effective zeros. The identified anomalous lateral acceleration data is then processed by applying Complete Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) to filter out anomalous components before signal reconstruction. The corrected data can then be used for subsequent analysis. The specific principle is as follows.

[0044] Figure 2 , Figure 3 These are, respectively, the abnormal vehicle body lateral acceleration data containing high-frequency interference components and the normal vehicle body lateral acceleration data in the embodiments of the present invention, for comparison. Figure 2 , Figure 3 It can be observed that, compared to the normal lateral acceleration of a vehicle body, the abnormal lateral acceleration curve exhibits a small-amplitude disturbance.

[0045] Figure 4 This is a magnified view of abnormal vehicle body lateral acceleration data containing high-frequency interference components in an embodiment of the present invention. Figure 4 As can be seen, the high-frequency interference exhibits a "sawtooth" pattern. Compared to normal vehicle lateral acceleration data, the abnormal vehicle lateral acceleration data containing high-frequency interference components shows greater fluctuations and stronger dispersion within a certain window length. Overall, the high-frequency interference caused by sensor anomalies only alters the local characteristics of vehicle lateral acceleration, without affecting the overall trend. In other words, the abnormal data is a superposition of normal vehicle lateral acceleration and abnormal high-frequency interference components.

[0046] Therefore, the identification of this type of abnormal lateral acceleration data of the vehicle body is transformed into: determining whether the lateral acceleration of the vehicle body has large fluctuations and strong dispersion within a small scale range.

[0047] In statistics, variance and standard deviation are used to measure data fluctuations. Considering that the amplitude of lateral acceleration of a vehicle body is generally within 0.03g, choosing standard deviation can better distinguish between normal and abnormal data. Obviously, the moving standard deviation of lateral acceleration data with local high-frequency spikes will be significantly greater than that of normal lateral acceleration data.

[0048] To better distinguish between abnormal and normal data numerically, the embodiments of this invention use moving standard deviation for quantitative description.

[0049] In one embodiment, calculating the moving standard deviation of the lateral acceleration sequence data of the car body in the turnout area to be processed includes:

[0050] Calculate the moving standard deviation of the lateral acceleration sequence data of the turnout area to be processed using the following formula:

[0051]

[0052]

[0053] In the formula, N is the total number of sampling points included in the lateral acceleration sequence data of the turnout area to be processed; i is the number of moves; j is the current sampling point number participating in the calculation within the calculation window; a i+j This represents the lateral acceleration amplitude of the j-th vehicle body within the window, which is calculated after the i-th movement. This represents the average value of the sampled points within the window used in the calculation after the i-th movement.

[0054] In this embodiment of the invention, the first threshold can be obtained through mathematical statistical analysis. For example, by comparing the moving standard deviation of 50 sets of vehicle body lateral acceleration data with local high-frequency spikes and 250 sets of normal vehicle body lateral acceleration data, the first threshold is set to 0.05g to distinguish between abnormal vehicle body lateral acceleration data containing local high-frequency components and normal vehicle body lateral acceleration data.

[0055] In this embodiment of the invention, the moving standard deviation was calculated for 28,000 vehicle lateral acceleration data points, each with 918 sampling points, and data exceeding 0.05g were extracted for result verification. It was found that in addition to data containing high-frequency interference, there was also a class of vehicle lateral acceleration data without high-frequency interference, exhibiting higher vibration frequencies (manifested as a large number of peaks and troughs). A typical example is shown below. Figure 5 As shown.

[0056] Figure 5 This is the original waveform of the lateral acceleration of the car body when the turnout passes through in an embodiment of the present invention. Figure 5 As can be seen, the lateral acceleration of the train body in this section contains multiple maximum and minimum points, but the amplitude of the lateral acceleration of the train body is moderate, indicating that when the EMU passes through this railway section, the lateral vibration frequency of the train body is high, but the vibration intensity is moderate.

[0057] Analysis based on the moving standard deviation calculation formula reveals that this type of vibration waveform is captured because of its high vibration frequency and large overall fluctuation in the lateral acceleration of the vehicle body within the segment. During calculation, the standard deviation value near the zero point is significant, contributing a large amplitude to the moving standard deviation. This results in the calculated moving standard deviation value being similar to that of abnormal vehicle body lateral acceleration data containing high-frequency interference components. Compared to abnormal vehicle body lateral acceleration data with localized high-frequency spikes, this type of highly fluctuating data is characterized by a higher number of zero points.

[0058] Therefore, in this embodiment of the invention, in order to improve the recognition accuracy, the number of zero points contained in the vehicle body lateral acceleration sequence is constrained.

[0059] Based on the above analysis, in this embodiment of the invention, the method for determining the anomaly of the lateral acceleration data of the vehicle body with local high-frequency burrs consists of two parts: (1) moving standard deviation and (2) number of effective zero points.

[0060] In one embodiment, calculating the number of valid zeros may include:

[0061] Find the position in the lateral acceleration sequence data of the car body in the turnout area to be processed where the amplitude of the lateral acceleration is 0. The sequence length of the lateral acceleration sequence data of the car body in the turnout area to be processed is, for example, N (N takes the range of 910 to 930).

[0062] Record the ordinal number of the position where the lateral acceleration amplitude of each car body is 0 in the lateral acceleration sequence data of the car body in the turnout area to be processed, and obtain a set containing M zero-point ordinal numbers:

[0063] {x1, x2, ..., x M}

[0064] In the formula, x M This indicates the ordinal number of the position where the Mth car body lateral acceleration amplitude is 0 in the lateral acceleration sequence data of the car body in the turnout area to be processed;

[0065] Using a set containing M zero-point ordinal numbers, the lateral acceleration sequence data of the turnout area to be processed is divided into multiple intervals, the interval range of which is represented as follows:

[0066]

[0067] Check sequentially whether the maximum value of the vehicle's lateral acceleration amplitude in the i-th interval exceeds the second threshold. If it does, record the i-th zero point as a valid zero point.

[0068] Count all valid zeros and output the number of valid zeros K.

[0069] In one embodiment, the second threshold is 0.005g and the third threshold is 20g.

[0070] Figure 6 This is a specific embodiment of the method for identifying and correcting local anomalies in the lateral acceleration of the car body in the turnout area according to an embodiment of the present invention, such as... Figure 6 As shown, the method includes:

[0071] Step 1: Obtain the lateral acceleration sequence data of the car body in the turnout area with a sequence length of N;

[0072] Step 2: Using formulas (1) and (2), calculate the moving standard deviation of the lateral acceleration sequence data of the turnout area with a sequence length of N;

[0073] Step 3: Determine if the moving standard deviation exceeds 0.05g. If it does, continue calculating the number of effective zeros.

[0074] Step 4: Calculate the number of significant zeros K;

[0075] Step 5: Determine the value of K. If K is less than 20, it is determined that there is high-frequency interference in the current vehicle lateral acceleration data.

[0076] Step 6: Correct the lateral acceleration of the vehicle body based on the CEEMDAN method.

[0077] In one embodiment, performing high-frequency interference filtering and signal reconstruction on the lateral acceleration sequence data of the car body in the turnout area to obtain the corrected lateral acceleration of the car body may include:

[0078] Perform CEEMDAN on the lateral acceleration sequence data of the car body in the turnout area to be processed, and obtain N sub-signals and 1 residual term;

[0079] Calculate the power spectral density of each of the N sub-signals;

[0080] Extract the dominant vibration frequency F of N sub-signals i ;

[0081] The dominant frequency F of the i-th sub-signal is determined sequentially. i If the frequency is greater than 20Hz, it is recorded as noise; if it is less than 20Hz, it is recorded as a valid signal.

[0082] By superimposing the valid signals, the corrected lateral acceleration of the vehicle body is obtained through reconstruction.

[0083] The effects and functions of this invention will be analyzed and illustrated below with an example:

[0084] Using the method for identifying and correcting local anomalies in the lateral acceleration of the car body in the turnout area in this embodiment of the invention, 28,000 sets of data were analyzed when the high-speed integrated inspection train passed through the turnout in a straight direction at a speed exceeding 100 km / h. A total of 120 sets of abnormal data were obtained, with an anomaly rate of 0.42%. This indicates that the vibration sensor used to collect the lateral acceleration of the car body during the operation of the high-speed integrated train is in a normal state for most of the operating conditions and most of the time.

[0085] Table 1 shows the standard deviation and effective zero-point count of the lateral acceleration of a train body in a certain turnout area. The numerical results all meet the conditions for the presence of high-frequency interference components, therefore it is determined to contain high-frequency interference components. The lateral acceleration waveform of the train body is as follows: Figure 7 As shown. Figure 7 This is a verification illustration of the method for identifying and correcting local anomalies in the lateral acceleration of the car body in the turnout area, as described in this embodiment of the invention. Figure 1 , Figure 7 The original waveform of the vehicle's lateral acceleration, which was identified as abnormal, is shown.

[0086] Table 1. Calculation of lateral acceleration index of a car body in a turnout area.

[0087] Moving standard deviation 0.5828 ≤0.50 yes Valid zeros 16 <20 yes

[0088] from Figure 7 It is also evident that the lateral acceleration data of the vehicle body does indeed contain high-frequency interference components. This verifies the effectiveness of the identification method proposed in this embodiment of the invention.

[0089] right Figure 7 The abnormal lateral acceleration data of the vehicle body shown is corrected using the correction method proposed in the embodiments of this invention. Figure 8 To correct the before-and-after comparison, Figure 8 This demonstrates a comparison of the vehicle's lateral acceleration before and after correction for the abnormal acceleration. From Figure 8 As can be seen, the data correction method proposed in the embodiments of the present invention can effectively eliminate high-frequency interference components contained in the original data.

[0090] This invention also provides a device for identifying and correcting local anomalies in the lateral acceleration of a car body in a turnout area, as described in the following embodiments. Since the principle behind this device is similar to the method for identifying and correcting local anomalies in the lateral acceleration of a car body in a turnout area, the implementation of this device can refer to the implementation of the method for identifying and correcting local anomalies in the lateral acceleration of a car body in a turnout area; repeated details will not be elaborated further.

[0091] Figure 9 This is a schematic diagram of the device for identifying and correcting local anomalies in the lateral acceleration of the car body in the turnout area, as described in an embodiment of the present invention. Figure 9 As shown, the device includes:

[0092] Anomaly identification module 901 is used to calculate the moving standard deviation of the lateral acceleration sequence data of the turnout area to be processed; compare the moving standard deviation with a first threshold; when the moving standard deviation exceeds the first threshold, calculate the number of valid zero points; wherein, the lateral acceleration sequence data of the turnout area to be processed is divided into multiple intervals, and the lateral acceleration amplitude at the two endpoints of any interval is 0, and the number of valid zero points is the total number of the intervals whose maximum lateral acceleration amplitude exceeds the second threshold before the first endpoint; when the number of valid zero points is lower than a third threshold, it is determined that there is high-frequency interference in the lateral acceleration sequence data of the turnout area to be processed.

[0093] The correction processing module 902 is used to perform high-frequency interference filtering and signal reconstruction on the lateral acceleration sequence data of the car body in the turnout area to obtain the corrected lateral acceleration of the car body.

[0094] In one embodiment, the anomaly detection module 901 is specifically used for:

[0095] Calculate the moving standard deviation of the lateral acceleration sequence data of the turnout area to be processed using the following formula:

[0096]

[0097]

[0098] In the formula, N is the total number of sampling points included in the lateral acceleration sequence data of the turnout area to be processed; i is the number of moves; j is the current sampling point number participating in the calculation within the calculation window; a i+j This represents the lateral acceleration amplitude of the j-th vehicle body within the window, which is calculated after the i-th movement. This represents the average value of the sampled points within the window used in the calculation after the i-th movement.

[0099] In one embodiment, the anomaly detection module 901 is specifically used for:

[0100] Locate the position in the lateral acceleration sequence data of the car body in the turnout area to be processed where the lateral acceleration amplitude of the car body is 0;

[0101] Record the ordinal number of the position where the lateral acceleration amplitude of each car body is 0 in the lateral acceleration sequence data of the car body in the turnout area to be processed, and obtain a set containing M zero-point ordinal numbers:

[0102] {x1, x2, ..., x M}

[0103] In the formula, x M This indicates the ordinal number of the position where the Mth car body lateral acceleration amplitude is 0 in the lateral acceleration sequence data of the car body in the turnout area to be processed;

[0104] Using a set containing M zero-point ordinal numbers, the lateral acceleration sequence data of the turnout area to be processed is divided into multiple intervals, the interval range of which is represented as follows:

[0105]

[0106] Check sequentially whether the maximum value of the vehicle's lateral acceleration amplitude in the i-th interval exceeds the second threshold. If it does, record the i-th zero point as a valid zero point.

[0107] Output the number of valid zeros.

[0108] In one embodiment, the correction processing module 902 is specifically used for:

[0109] Perform CEEMDAN on the lateral acceleration sequence data of the car body in the turnout area to obtain N sub-signals and 1 residual term;

[0110] Calculate the power spectral density of each of the N sub-signals;

[0111] Extract the dominant vibration frequency F of N sub-signals i ;

[0112] The dominant frequency F of the i-th sub-signal is determined sequentially. i If the frequency is greater than 20Hz, it is recorded as noise; if it is less than 20Hz, it is recorded as a valid signal.

[0113] By superimposing the valid signals, the corrected lateral acceleration of the vehicle body is obtained through reconstruction.

[0114] This invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-mentioned method for identifying and correcting local abnormal data of lateral acceleration of the vehicle body in the turnout area.

[0115] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-mentioned method for identifying and correcting local anomalies in the lateral acceleration of the vehicle body in the turnout area.

[0116] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-mentioned method for identifying and correcting local abnormal data of lateral acceleration of the vehicle body in the turnout area.

[0117] In this embodiment of the invention, the moving standard deviation of the lateral acceleration sequence data of the car body in the turnout area to be processed is calculated; the moving standard deviation is compared with a first threshold; when the moving standard deviation exceeds the first threshold, the number of valid zeros is calculated; wherein, the lateral acceleration sequence data of the car body in the turnout area to be processed is divided into multiple intervals, and the lateral acceleration amplitude at the two endpoints of any interval is 0, and the number of valid zeros is the total number of the intervals whose maximum lateral acceleration amplitude exceeds the second threshold before the first endpoint; when the number of valid zeros is lower than a third threshold, it is determined that there is high-frequency interference in the lateral acceleration sequence data of the car body in the turnout area to be processed; high-frequency interference filtering and signal reconstruction are performed on the lateral acceleration sequence data of the car body in the turnout area to be processed to obtain the corrected lateral acceleration of the car body. In this embodiment of the invention, the moving standard deviation of the lateral acceleration sequence data of the car body in the turnout area to be processed is first calculated to determine whether the lateral acceleration sequence data of the car body in the turnout area to be processed is abnormal data and has the possibility of high-frequency interference. When the moving standard deviation result exceeds the first threshold, it indicates that the lateral acceleration sequence data of the car body in the turnout area to be processed is abnormal data and has a high possibility of high-frequency interference. The effective number of zero points is further calculated. When the effective number of zero points is lower than the third threshold, it indicates that the lateral acceleration sequence data of the car body in the turnout area to be processed has high-frequency interference. This method effectively improves the accuracy of identifying abnormal lateral acceleration data of the car body, thereby correcting the lateral acceleration sequence data of the car body in the turnout area to be processed, and ensuring the validity and reliability of the results obtained when analyzing the lateral acceleration data of the car body.

[0118] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied 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.

[0119] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0120] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0121] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0122] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are 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 within the scope of protection of the present invention.

Claims

1. A method for identifying and correcting local anomalies in the lateral acceleration data of a train body in a turnout area, characterized in that, include: Calculate the moving standard deviation of the lateral acceleration sequence data of the car body in the turnout area to be processed; Compare the moving standard deviation with the first threshold; When the moving standard deviation result exceeds the first threshold, the number of valid zero points is calculated; wherein, the lateral acceleration sequence data of the turnout area to be processed is divided into multiple intervals, the lateral acceleration amplitude at the two endpoints of any interval is 0, and the number of valid zero points is the total number of the intervals whose maximum lateral acceleration amplitude exceeds the second threshold before the first endpoint. When the number of valid zero points is lower than the third threshold, it is determined that there is high-frequency interference in the lateral acceleration sequence data of the turnout area to be processed. High-frequency interference filtering and signal reconstruction are performed on the lateral acceleration sequence data of the car body in the turnout area to obtain the corrected lateral acceleration of the car body. The calculation of the number of significant zeros includes: Locate the position in the lateral acceleration sequence data of the car body in the turnout area to be processed where the lateral acceleration amplitude of the car body is 0; Record the ordinal number of the position where the lateral acceleration amplitude of each car body is 0 in the lateral acceleration sequence data of the car body in the turnout area to be processed, and obtain the data containing... M A set of zero-point ordinal numbers: In the formula, Indicates the first M The ordinal number of the position where the lateral acceleration amplitude of the car body is 0 in the lateral acceleration sequence data of the car body in the turnout area to be processed; Using inclusion M The set of zero-point ordinal numbers divides the lateral acceleration sequence data of the car body in the turnout area to be processed into multiple intervals, and the interval range is represented as follows: Judge the first one in sequence i Does the maximum value of the vehicle's lateral acceleration amplitude in each interval exceed the second threshold? If it does, record the value as... i Each zero point is a valid zero point; Output the number of valid zeros.

2. The method as described in claim 1, characterized in that, Calculate the moving standard deviation of the lateral acceleration sequence data of the car body in the turnout area to be processed, including: Calculate the moving standard deviation of the lateral acceleration sequence data of the turnout area to be processed using the following formula: In the formula, N This represents the total number of sampling points included in the lateral acceleration sequence data of the car body in the turnout area to be processed; i The number of moves; j This is the index of the sampling point currently participating in the calculation within the calculation window. For the first i The window that participates in the calculation after the next movement j The lateral acceleration amplitude of the vehicle body; For the first i The average value of the sampling points within the window used in the calculation after each movement.

3. The method as described in claim 1, characterized in that, The lateral acceleration sequence data of the car body in the turnout area to be processed is subjected to high-frequency interference filtering and signal reconstruction to obtain the corrected lateral acceleration of the car body, including: Performing an adaptive noise complete set empirical mode decomposition (CEEMDAN) on the lateral acceleration sequence data of the car body in the turnout area to obtain... N One sub-signal and one residual term; Calculate separately N The power spectral density of each sub-signal; extract N The main frequency of each sub-signal vibration F i ; Judge the first one in sequence i The main frequency of each sub-signal vibration F i If the frequency is greater than 20Hz, it is recorded as noise; if it is less than 20Hz, it is recorded as a valid signal. By superimposing the valid signals, the corrected lateral acceleration of the vehicle body is obtained through reconstruction.

4. A device for identifying and correcting local anomalies in the lateral acceleration of a train body in a turnout area, characterized in that, include: The anomaly detection module is used to calculate the moving standard deviation of the lateral acceleration sequence data of the car body in the turnout area to be processed; Compare the moving standard deviation with the first threshold; When the moving standard deviation exceeds the first threshold, the number of valid zeros is calculated. The lateral acceleration sequence data of the turnout area is divided into multiple intervals, with the lateral acceleration amplitude at both endpoints of any interval being 0. The number of valid zeros is the total number of intervals whose maximum lateral acceleration amplitude exceeds the second threshold, preceding the first endpoint. When the number of valid zeros is below the third threshold, it is determined that the lateral acceleration sequence data of the turnout area contains high-frequency interference. The correction processing module is used to perform high-frequency interference filtering and signal reconstruction on the lateral acceleration sequence data of the car body in the turnout area to obtain the corrected lateral acceleration of the car body. Specifically, the anomaly detection module is used for: Locate the position in the lateral acceleration sequence data of the car body in the turnout area to be processed where the lateral acceleration amplitude of the car body is 0; Record the ordinal number of the position where the lateral acceleration amplitude of each car body is 0 in the lateral acceleration sequence data of the car body in the turnout area to be processed, and obtain the data containing... M A set of zero-point ordinal numbers: In the formula, Indicates the first M The ordinal number of the position where the lateral acceleration amplitude of the car body is 0 in the lateral acceleration sequence data of the car body in the turnout area to be processed; Using inclusion M The set of zero-point ordinal numbers divides the lateral acceleration sequence data of the car body in the turnout area to be processed into multiple intervals, and the interval range is represented as follows: Judge the first one in sequence i Does the maximum value of the vehicle's lateral acceleration amplitude in each interval exceed the second threshold? If it does, record the value as... i Each zero point is a valid zero point; Output the number of valid zeros.

5. The apparatus as described in claim 4, characterized in that, The anomaly detection module is specifically used for: Calculate the moving standard deviation of the lateral acceleration sequence data of the turnout area to be processed using the following formula: In the formula, N This represents the total number of sampling points included in the lateral acceleration sequence data of the car body in the turnout area to be processed; i The number of moves; j This is the index of the sampling point currently participating in the calculation within the calculation window. For the first i The window that participates in the calculation after the next movement j The lateral acceleration amplitude of the vehicle body; For the first i The average value of the sampling points within the window used in the calculation after each movement.

6. The apparatus as claimed in claim 4, characterized in that, The correction processing module is specifically used for: The adaptive noise complete set empirical mode decomposition (CEEMDAN) is performed on the lateral acceleration sequence data of the car body in the turnout area to obtain... N One sub-signal and one residual term; Calculate separately N The power spectral density of each sub-signal; extract N The main frequency of each sub-signal vibration F i ; Judge the first one in sequence i The main frequency of each sub-signal vibration F i If the frequency is greater than 20Hz, it is recorded as noise; if it is less than 20Hz, it is recorded as a valid signal. By superimposing the valid signals, the corrected lateral acceleration of the vehicle body is obtained through reconstruction.

7. 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, it implements the method of any one of claims 1 to 3.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method of any one of claims 1 to 3.

9. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the method of any one of claims 1 to 3.

Images