Rail transit system fault detection method and device

Abstract

The invention discloses a rail transit system fault detection method and device, and the method comprises the steps: carrying out the filtering of a vibration signal according to the target frequency bands of different to-be-detected positions in a rail transit system, so as to generate target vibration signals of different to-be-detected positions; as the fault characteristic amplitudes of the different to-be-detected positions are different from the fault characteristic standard, when the fault characteristic amplitudes of the to-be-detected positions conform to the corresponding fault characteristic standard of the to-be-detected positions, the fault characteristic amplitudes of the to-be-detected positions can be detected, and the fault characteristic amplitudes of the to-be-detected positions can be detected according to the fault characteristic standard of the to-be-detected positions. And determining that the corresponding to-be-detected position has a fault. On the basis, the fault characteristic amplitude representing whether the multiple to-be-detected positions fail or not is determined from one vibration signal only by collecting one vibration signal, so that fault detection of the multiple to-be-detected positions is achieved, the fault detection efficiency of the rail transit system is improved, and the fault detection cost is reduced.

Description

Technical Field

[0001] This invention relates to the field of rail transit system testing, and in particular to a method and apparatus for fault detection in rail transit systems. Background Technology

[0002] As a means of transportation carrying passengers and goods, the safe operation of rail transit systems is crucial to the lives of various people. Therefore, it is necessary to test whether rail transit systems can operate normally and safely. Furthermore, due to the current trend of developing rail transit systems towards intensification and lightweighting, the demand for integrated fault detection of different components of rail transit systems is increasing. Based on this, how to efficiently and accurately test rail transit systems is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0003] The purpose of this invention is to provide a fault detection method and apparatus for rail transit systems. By collecting only one vibration signal, the fault characteristic amplitude that characterizes whether multiple test locations are faulty can be determined from the vibration signal, thereby realizing fault detection at multiple test locations, improving the fault detection efficiency of rail transit systems, and reducing the fault detection cost.

[0004] To solve the above-mentioned technical problems, the present invention provides a fault detection method for rail transit systems, comprising:

[0005] The vibration signal at the axle box of the vehicle is acquired, and the vibration signal is filtered for the target frequency band corresponding to each test location in the rail transit system to determine the target vibration signal of the target frequency band corresponding to each test location.

[0006] Spectral analysis is performed on the target vibration signal corresponding to each of the measured positions to obtain the spectral information of each target vibration signal;

[0007] The fault characteristic amplitude of each of the test locations is obtained from the spectral information of each of the target vibration signals;

[0008] Determine whether the fault characteristic amplitude of each of the locations to be tested conforms to the corresponding fault characteristic standard;

[0009] If the fault characteristic amplitude meets the fault characteristic standard of the corresponding test location, then it is determined that there is a fault at the corresponding test location.

[0010] Preferably, the locations to be tested include bearings, wheels, and rails;

[0011] The vibration signal is filtered according to the target frequency band corresponding to each test location in the rail transit system to determine the target vibration signal of each test location in the target frequency band, including:

[0012] Determine the first minimum fault modulation frequency when the bearing has a fault, the first maximum fault modulation frequency when the wheel has a fault, and the second minimum fault modulation frequency and the second maximum fault modulation frequency when the rail has a fault.

[0013] The vibration signal is high-pass filtered based on the first lowest fault modulation frequency to determine the first target vibration signal corresponding to the bearing with a frequency higher than the first lowest fault modulation frequency.

[0014] The vibration signal is low-pass filtered based on the first highest fault modulation frequency to determine the second target vibration signal corresponding to the wheel with a frequency lower than the first highest fault modulation frequency.

[0015] The vibration signal is bandpass filtered based on the second lowest fault modulation frequency and the second highest fault modulation frequency to determine a third target vibration signal whose frequency corresponds to the rail in the target frequency band between the second lowest fault modulation frequency and the second highest fault modulation frequency.

[0016] Preferably, the fault characteristic amplitude of each of the measured locations is obtained from the spectral information of each of the target vibration signals, including:

[0017] The first fault characteristic amplitude of the bearing is obtained from the spectral information of the first target vibration signal;

[0018] The second fault characteristic amplitude of the wheel is obtained from the spectral information of the second target vibration signal;

[0019] The third fault characteristic amplitude of the rail is obtained from the spectral information of the third target vibration signal;

[0020] Determine whether the fault characteristic amplitude at each of the locations to be tested conforms to the corresponding fault characteristic standard, including:

[0021] Determine whether the amplitude of the first fault characteristic of the bearing meets the fault characteristic standard of the bearing;

[0022] Determine whether the second fault characteristic amplitude of the wheel meets the fault characteristic standard of the wheel;

[0023] Determine whether the amplitude of the third fault characteristic of the rail meets the fault characteristic standard of the rail.

[0024] Preferably, before obtaining the first fault characteristic amplitude of the bearing from the spectral information of the first target vibration signal, the method further includes:

[0025] The fault characteristic frequencies of each component under test of the bearing are determined based on the bearing parameters and the bearing's operating conditions.

[0026] Obtaining the first fault characteristic amplitude of the bearing from the spectral information of the first target vibration signal includes:

[0027] The first fault characteristic amplitude values ​​of the fault characteristic frequencies of each component under test of the bearing, from the first target vibration signal spectrum information, are obtained from the first harmonic to the Mth harmonic of the fault characteristic frequency.

[0028] Preferably, determining whether the amplitude of the first fault characteristic of the bearing meets the fault characteristic standard of the bearing includes:

[0029] Calculate the average fault characteristic amplitude of the first fault characteristic amplitude from the first harmonic to the Mth harmonic of the fault characteristic frequency of each component under test of the bearing;

[0030] Determine the alarm limit values ​​corresponding to each component under test of the bearing;

[0031] If the average fault characteristic amplitude is greater than the alarm limit of the corresponding component under test, then the first fault characteristic amplitude of the bearing is determined to meet the fault characteristic standard of the bearing.

[0032] Preferably, after calculating the average fault characteristic amplitude of the first fault characteristic amplitude from the first harmonic to the Mth harmonic of the fault characteristic frequency of each component under test of the bearing, the method further includes:

[0033] Determine the warning limit for each component under test of the bearing; the alarm limit is greater than the warning limit.

[0034] If the average fault characteristic amplitude is greater than the warning limit of the corresponding component under test, but not greater than the alarm limit of the corresponding component under test, then a fault warning is issued for the corresponding component under test.

[0035] If the average fault characteristic amplitude is not greater than the warning limit of the corresponding component under test, then it is determined that the corresponding component under test does not have a fault.

[0036] Preferably, obtaining the second fault characteristic amplitude of the wheel from the spectral information of the second target vibration signal includes:

[0037] The maximum value of the current frequency domain signal in the spectral information of the second target vibration signal is determined, and the maximum value of the current frequency domain signal is determined as the second fault characteristic amplitude of the wheel.

[0038] Preferably, determining whether the second fault characteristic amplitude of the wheel meets the fault characteristic standard of the wheel includes:

[0039] Determine the target fault characteristic frequency corresponding to the second fault characteristic amplitude from the spectral information of the second target vibration signal;

[0040] Determine the rotational frequency of the wheel;

[0041] The effective order of the polygonal fault of the wheel is determined based on the ratio between the target fault characteristic frequency and the rotational frequency.

[0042] Obtain and determine the N second fault feature amplitudes that are continuously determined before the current time as N fault feature amplitudes to be measured;

[0043] Obtain and determine the N valid orders that are consecutively determined before the current time as the N valid orders to be tested;

[0044] If the amplitude values ​​of all N fault features to be tested are greater than the wheel alarm limit, and the N valid orders to be tested are the same, then the second fault feature amplitude value of the wheel is determined to meet the fault feature standard of the wheel.

[0045] Preferably, after acquiring and determining the N valid orders consecutively determined before the current time as the N valid orders to be measured, the method further includes:

[0046] If the amplitude values ​​of the N fault features to be tested are not all greater than the wheel alarm limit, but are all greater than the wheel warning limit, and the N valid orders to be tested are the same, then a wheel fault warning is issued.

[0047] If the amplitude of all N fault characteristics to be tested is not greater than the wheel warning limit, and the N valid orders to be tested are the same, then it is determined that the wheel does not have a fault.

[0048] Preferably, determining the effective order of the polygonal fault of the wheel based on the ratio between the target fault characteristic frequency and the rotational frequency includes:

[0049] The ratio between the target fault characteristic frequency and the rotational frequency is determined as the calculation order of the polygonal fault of the wheel;

[0050] The calculated order is rounded down to determine the theoretical order;

[0051] The order deviation is determined based on the absolute value of the difference between the calculated order and the theoretical order;

[0052] If the order deviation is less than the deviation limit, then the theoretical order is determined as the effective order;

[0053] If the order deviation is not less than the deviation limit, then the maximum value of the frequency domain signal other than the current maximum value of the frequency domain signal in the spectrum information of the second target vibration signal is updated to the current maximum value of the frequency domain signal, and the step of determining the current maximum value of the frequency domain signal as the second fault characteristic amplitude of the wheel is returned.

[0054] Preferably, obtaining the third fault characteristic amplitude of the rail from the spectral information of the third target vibration signal includes:

[0055] The maximum value and effective value of the rail frequency domain signal are obtained from the spectral information of the vibration signal of the third target.

[0056] Preferably, determining whether the third fault characteristic amplitude of the rail meets the fault characteristic standard of the rail includes:

[0057] The maximum value of the rail frequency domain signal and the average value of the rail frequency domain signal are determined as rail fault characteristic indicators.

[0058] Determine the total time period during which all carriages of the vehicle pass the target test point on the rail;

[0059] Obtain the rail fault characteristic indicators of each car of the vehicle within the total time period.

[0060] The maximum value of the rail fault characteristic index corresponding to each of the various rail fault characteristic indices of each of the various rail fault characteristic indices within the total time period is determined from each of the rail fault characteristic indices of each of the various ... indices.

[0061] If the maximum value of the rail fault characteristic index corresponding to each of the carriages is greater than the rail fault alarm limit, then the third fault characteristic amplitude of the rail is determined to meet the rail fault characteristic standard.

[0062] Preferably, after determining the maximum value of the rail fault characteristic index corresponding to each of the carriages from the rail fault characteristic indexes within the total time period, the method further includes:

[0063] If the maximum value of the rail fault characteristic index corresponding to each of the carriages is not greater than the rail fault alarm limit, but is greater than the rail fault warning limit, then a rail fault warning is issued.

[0064] If the maximum value of the rail fault characteristic index corresponding to each of the carriages is not greater than the rail fault warning limit, then it is determined that the rail has no fault.

[0065] To solve the above-mentioned technical problems, the present invention provides a fault detection device for a rail transit system, comprising:

[0066] Memory, used to store computer programs;

[0067] A processor is used to implement the steps of the fault detection method for rail transit systems as described above when executing a computer program.

[0068] This application provides a fault detection method and apparatus for rail transit systems. In this scheme, vibration signals are filtered according to the target frequency bands of different test locations within the rail transit system to generate target vibration signals for different test locations. Fault characteristic amplitudes for different test locations are then determined based on these different target vibration signals. Since the fault characteristic amplitudes for different test locations differ from the fault characteristic standards, a fault can be determined at a test location when the fault characteristic amplitude matches the corresponding fault characteristic standard. Based on this, this application only acquires a single vibration signal and determines the fault characteristic amplitudes representing whether multiple test locations are faulty from this single signal, thereby achieving fault detection for multiple test locations, improving the fault detection efficiency of rail transit systems, and reducing fault detection costs. Attached Figure Description

[0069] To more clearly illustrate the embodiments of the present invention, the accompanying drawings used in the embodiments 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.

[0070] Figure 1 A flowchart illustrating a fault detection method for a rail transit system provided in this application;

[0071] Figure 2 This application provides a structural schematic diagram of a fault detection system for a rail transit system.

[0072] Figure 3 A schematic diagram of the structure of a fault detection device for a rail transit system provided in this application;

[0073] Figure 4 This is a schematic diagram of the structure of a computer-readable storage medium provided in this application. Detailed Implementation

[0074] The core of this invention is to provide a fault detection method and device for rail transit systems. By collecting only one vibration signal, the fault characteristic amplitude that characterizes whether multiple test locations are faulty can be determined from the vibration signal, thereby realizing fault detection at multiple test locations, improving the fault detection efficiency of rail transit systems, and reducing the fault detection cost.

[0075] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0076] Please refer to Figure 1 , Figure 1 This application provides a flowchart illustrating a fault detection method for a rail transit system, the method comprising:

[0077] S11: Acquire the vibration signal at the axle box of the vehicle, and filter the vibration signal for the target frequency band corresponding to each test location in the rail transit system to determine the target vibration signal of the target frequency band corresponding to each test location.

[0078] During the operation of a rail transit system, fault detection at every location of the vehicle is crucial to the safe operation of the rail transit system.

[0079] In this application, during the fault detection process of the rail transit system, the vibration signal at the axle box of the vehicle is first acquired. In order to detect different test locations of the rail transit system according to the vibration signal, the vibration signal is filtered according to the target frequency band in the vibration signal corresponding to the location of the test location, so as to determine the target vibration signal of the target frequency band required for fault detection of different test locations.

[0080] For example, different test locations include bearings, wheels, and rails. When a bearing fails, the vibration signal corresponds to the bearing's fault signal in the target frequency band. Therefore, the vibration signal can be filtered according to the bearing's target frequency band to determine the target vibration signal, thus identifying whether the bearing has failed. Similarly, when a wheel fails, the vibration signal corresponds to the wheel's fault signal in the target frequency band. Filtering the vibration signal according to the wheel's target frequency band helps determine the wheel's fault signal, thus identifying whether the wheel has failed. Likewise, when a rail fails, the vibration signal corresponds to the rail's fault signal in the target frequency band. Filtering the vibration signal according to the rail's target frequency band helps determine the rail's fault signal, thus identifying whether the rail has failed.

[0081] S12: Perform spectrum analysis on the target vibration signal corresponding to each test position to obtain the spectrum information of each target vibration signal;

[0082] After determining the target vibration signal corresponding to each test location, when performing fault detection on different test locations based on different target vibration signals, the spectrum analysis of each target vibration signal is first performed to obtain the spectrum information of each target vibration signal.

[0083] In the process of spectral analysis of the target vibration signal, the Hilbert transform is first used to demodulate the envelope of the target vibration signal. For the Hilbert transform, let the given signal, that is, the target vibration signal, be x(t), its Hilbert transform is defined as follows:

[0084] ;

[0085] in, Let x(t) be an estimate, where t is time. As the integral variable, it can be seen that after the Hilbert transform, It is 90° out of phase with x(t). If x(t) is taken as the real part, The imaginary part can be used to construct an analytic signal. Therefore, the envelope curve equation of the target vibration signal can be obtained:

[0086] ;

[0087] Perform a Fourier transform on the above expression to obtain the spectral information of the target vibration signal.

[0088] S13: Obtain the fault characteristic amplitude of each test location from the spectral information of each target vibration signal;

[0089] Impact alters the frequency structure of the target vibration signal at the test location, manifesting as specific frequency components in the frequency domain, such as fault characteristic frequencies. By performing spectral analysis on the target vibration signal, the spectral information of the target vibration signal can be determined. Then, different fault characteristic frequencies corresponding to different fault types can be identified from the spectral information. Each fault characteristic frequency is found in the spectral information, and the fault characteristic amplitude of each fault characteristic frequency is determined. In the spectral information, the fault characteristic frequency is on the horizontal axis, while the fault characteristic amplitude is the vertical axis corresponding to the fault characteristic frequency.

[0090] The fault characteristic frequencies differ at different test locations. These frequencies are present in the target vibration signals at those locations. To determine the fault characteristic amplitude at each test location, it is necessary to identify the corresponding fault characteristic frequencies from the target vibration signals at those locations, and then determine the fault characteristic amplitude based on these frequencies. For example, the fault characteristic frequency of a bearing failure can be determined from the target vibration signal of a bearing, and the fault characteristic amplitude can be determined based on this frequency. Similarly, the fault characteristic frequency of a wheel failure can be determined from the target vibration signal of a wheel, and the fault characteristic amplitude can be determined based on this frequency. Likewise, the fault characteristic frequency of a rail failure can be determined from the target vibration signal of a rail, and the fault characteristic amplitude can be determined based on this frequency.

[0091] S14: Determine whether the fault characteristic amplitude of each test location meets the corresponding fault characteristic standard;

[0092] After determining the fault characteristic amplitudes at different test locations, the fault characteristic amplitudes are analyzed according to the fault characteristic standards for different test locations to determine whether the fault characteristic amplitudes at different test locations meet the fault characteristic standards corresponding to that test location.

[0093] Among them, the fault characteristic standard refers to the standard that reflects the fault when a fault occurs at the location under test.

[0094] S15: If the fault characteristic amplitude meets the fault characteristic standard of the corresponding test location, then it is determined that there is a fault at the corresponding test location.

[0095] If the corresponding fault characteristic amplitude meets the fault characteristic standard of the corresponding test location, and there is a fault at the test location, for example, if the fault characteristic amplitude of the bearing meets the fault characteristic standard when the bearing fails, then it is determined that the bearing has a fault; if the fault characteristic amplitude of the wheel meets the fault characteristic standard when the wheel fails, then it is determined that the wheel has a fault; if the fault characteristic amplitude of the rail meets the fault characteristic standard when the rail fails, then it is determined that the rail has a fault.

[0096] In summary, this application decouples the vibration signal, thereby acquiring only one vibration signal and determining the fault characteristic amplitude from it to characterize whether multiple test locations are faulty. This enables fault detection at multiple test locations, improving the fault detection efficiency of rail transit systems and reducing fault detection costs.

[0097] Based on the above embodiments:

[0098] In one preferred embodiment, the locations to be measured include bearings, wheels, and rails;

[0099] Vibration signals are filtered according to the target frequency bands corresponding to each test location in the rail transit system to determine the target vibration signal for each test location in the target frequency band, including:

[0100] Determine the first minimum fault modulation frequency when the bearing has a fault, the first maximum fault modulation frequency when the wheel has a fault, and the second minimum fault modulation frequency and the second maximum fault modulation frequency when the rail has a fault.

[0101] The vibration signal is high-pass filtered based on the first lowest fault modulation frequency to determine the first target vibration signal corresponding to the bearing in the target frequency band that is higher than the first lowest fault modulation frequency.

[0102] The vibration signal is low-pass filtered based on the first highest fault modulation frequency to determine the second target vibration signal corresponding to the wheel in the target frequency band that is lower than the first highest fault modulation frequency.

[0103] The vibration signal is bandpass filtered based on the second lowest fault modulation frequency and the second highest fault modulation frequency to determine the third target vibration signal whose frequency corresponds to the rail in the target frequency band between the second lowest fault modulation frequency and the second highest fault modulation frequency.

[0104] In this embodiment, the locations to be tested are specifically divided into bearings, wheels, and rails. Bearings and wheels are important load-bearing components of rail transit vehicles. Wheels and rails are in direct contact. The vibration signals collected at the axle box in this application can simultaneously reflect the health status of bearings, wheels, and rails. Therefore, a vibration sensor is installed at the axle box to collect vibration signals, thereby achieving the effect of detecting faults at multiple locations with a single sensor. Bearing faults include, but are not limited to, peeling, electrolytic corrosion, and abrasion; wheel faults include, but are not limited to, abrasion, peeling, and wheel polygonal structures; and rail faults include, but are not limited to, track irregularities, track corrugation, and oil contamination.

[0105] To perform fault detection on bearings, wheels, and rails separately based on the same vibration signal, this embodiment determines the target frequency band for each test location based on the fault modulation frequency at different test locations. For example, when a bearing has a fault, the first lowest fault modulation frequency characterizes the fault; when a wheel has a fault, the first highest fault modulation frequency characterizes the fault; and when a rail has a fault, the second lowest fault modulation frequency and the second highest fault modulation frequency characterize the fault. Based on this, the vibration signal is high-pass filtered according to the first lowest fault modulation frequency. The target frequency band corresponding to the bearing is a frequency band higher than the first lowest fault modulation frequency, and the first target vibration signal corresponding to the bearing is a high-frequency band. The vibration signal is at the first lowest fault modulation frequency; the vibration signal is low-pass filtered based on the first highest fault modulation frequency, the target frequency band corresponding to the wheel is the frequency band lower than the first highest fault modulation frequency, and the second target vibration signal corresponding to the wheel is the target vibration signal with a frequency lower than the first highest fault modulation frequency; the vibration signal is band-pass filtered according to the second lowest fault modulation frequency and the second highest fault modulation frequency, the target frequency band corresponding to the rail is the frequency band higher than the second lowest fault modulation frequency and lower than the second highest fault modulation frequency, and the third target vibration signal corresponding to the rail is the vibration signal with a frequency higher than the second lowest fault modulation frequency and lower than the second highest fault modulation frequency.

[0106] The modulation frequencies of the first highest fault, the second lowest fault, and the second highest fault are not limited in magnitude. The first lowest fault modulation frequency can be between 2000Hz and 4000Hz, and the first highest fault modulation frequency can also be between 2000Hz and 4000Hz. The second lowest fault modulation frequency and the second highest fault modulation frequency are determined by statistically analyzing the wavelengths of rail corrugation occurring in the rail transit system lines. The upper and lower limits of the rail corrugation wavelengths are then determined, and the second lowest fault modulation frequency and the second highest fault modulation frequency are determined based on these limits. The second lowest fault modulation frequency = vehicle speed / upper rail corrugation wavelength, and the second highest fault modulation frequency = vehicle speed / lower rail corrugation wavelength.

[0107] As a preferred embodiment, the fault characteristic amplitude of each test location is obtained from the spectral information of each target vibration signal, including:

[0108] The first fault characteristic amplitude of the bearing is obtained from the spectral information of the first target vibration signal;

[0109] The second fault characteristic amplitude of the wheel is obtained from the spectral information of the second target vibration signal;

[0110] The amplitude of the third fault characteristic of the rail is obtained from the spectral information of the vibration signal of the third target.

[0111] Determine whether the fault characteristic amplitude at each test location meets the corresponding fault characteristic standard, including:

[0112] Determine whether the amplitude of the first fault characteristic of the bearing meets the bearing fault characteristic standard;

[0113] Determine whether the amplitude of the second fault characteristic of the wheel meets the fault characteristic standard of the wheel;

[0114] Determine whether the amplitude of the third fault characteristic of the rail meets the fault characteristic standard of the rail.

[0115] In order to detect faults in bearings, wheels, and rails respectively based on the first target vibration signal, the second target vibration signal, and the third target vibration signal, fault characteristic standards corresponding to bearings, wheels, and rails are also determined respectively, so as to more accurately determine whether there is a fault at each test location based on the decoupled vibration signal.

[0116] As a preferred embodiment, before obtaining the first fault characteristic amplitude of the bearing from the spectral information of the first target vibration signal, the method further includes:

[0117] The fault characteristic frequencies of each component under test in the bearing are determined based on the bearing parameters and the bearing's operating conditions.

[0118] The first fault characteristic amplitude of the bearing is obtained from the spectral information of the first target vibration signal, including:

[0119] The first fault characteristic amplitude values ​​from the first harmonic to the Mth harmonic of the fault characteristic frequency of each component under test of the bearing are obtained from the spectral information of the first target vibration signal.

[0120] When determining whether a bearing is faulty, the fault characteristic frequency of each component of the bearing can be calculated using the bearing parameters and operating conditions. By performing fault detection on each component of the bearing, it can be determined whether the bearing is faulty. When any component of the bearing is faulty, it indicates that the bearing is faulty.

[0121] Among them, the failure characteristic frequency of the bearing outer ring for:

[0122] ;

[0123] Fault characteristic frequency of bearing inner ring for:

[0124] ;

[0125] Fault characteristic frequency of bearing rolling elements for:

[0126] ;

[0127] Fault characteristic frequency of bearing conversion frame for:

[0128] ;

[0129] Where D is the bearing pitch circle diameter, specifically the diameter of the circle containing the center of the rolling elements, and d is the diameter of the rolling element. η is the contact angle, which physically represents the angle between the direction of the force applied to the rolling element and the perpendicular lines of the inner and outer raceways. n is the number of rolling elements. This is the rotational frequency of the shaft where the bearing is located.

[0130] The corresponding fault characteristic amplitude can be extracted from the frequency of the fault characteristic of different bearing components in the spectrum information of the first target vibration signal.

[0131] Furthermore, since the faults of various bearing components are not only characterized at the corresponding fault characteristic frequency, but may also be characterized at the harmonics of the fault characteristic frequency, that is, there will also be a corresponding increase in amplitude at the harmonics of the fault characteristic frequency, the amplitude of 1 to M times the fault characteristic frequency of each bearing component can be selected as the fault characteristic amplitude of that component of the bearing. For ease of calculation, M can be an integer less than 7.

[0132] As a preferred embodiment, determining whether the amplitude of the first fault characteristic of the bearing meets the bearing's fault characteristic criteria includes:

[0133] Calculate the average fault characteristic amplitude of the first fault characteristic amplitude from the first harmonic to the Mth harmonic of the fault characteristic frequency of each component under test of the bearing;

[0134] Determine the alarm limit values ​​corresponding to each component under test in the bearing;

[0135] If the average fault characteristic amplitude is greater than the alarm limit of the corresponding component under test, then the first fault characteristic amplitude of the bearing is determined to meet the bearing fault characteristic standard.

[0136] By calculating the average fault characteristic amplitude at different harmonic frequencies, the fault characteristic indicators corresponding to different components of the bearing can be obtained. For example, for the outer ring of the bearing: , The average fault characteristic amplitude of the outer ring. These represent the frequencies from 1st to Mth harmonics for the outer ring; for the inner ring of the bearing: , The average fault characteristic amplitude of the inner circle. These represent the frequencies from 1st to Mth harmonics for the inner ring; for the rolling elements of the bearing: , This represents the average fault characteristic amplitude of the rolling element. These represent the frequencies from 1st to Mth harmonics of the rolling elements; for the bearing cage: , This represents the average fault characteristic amplitude of the rolling element. These represent the 1st to Mth harmonics of the cage.

[0137] As a preferred embodiment, after calculating the average fault characteristic amplitude of the first fault characteristic amplitude from the first harmonic to the Mth harmonic of the fault characteristic frequency of each component under test in the bearing, the method further includes:

[0138] Determine the warning limits for each component under test in the bearing; the alarm limit is greater than the warning limit;

[0139] If the average fault characteristic amplitude is greater than the warning limit of the corresponding component under test, but not greater than the alarm limit of the corresponding component under test, then a fault warning is issued for the corresponding component under test.

[0140] If the average fault characteristic amplitude is not greater than the warning limit of the corresponding component under test, then it is determined that the corresponding component under test does not have a fault.

[0141] Set warning limits F1 and alarm limits F2 for different fault characteristic indicators of different components under test in the bearing. The warning limits F1 and alarm limits F2 are different for different components under test in the bearing. First, determine whether the average fault characteristic amplitude of each component under test in the bearing exceeds the alarm limit F2. If it exceeds F2, an alarm is reported for the corresponding component under test in the bearing. If the average fault characteristic amplitude does not exceed the alarm limit F2, determine whether the average fault characteristic amplitude exceeds the warning limit F1. If the average fault characteristic amplitude exceeds the warning limit F1, a warning is reported for the corresponding component under test in the bearing; otherwise, the corresponding component under test in the bearing is reported as normal.

[0142] It should be noted that the warning limit F1 for the inner ring, outer ring, rolling elements, and cage of the bearing is not the same, and the alarm limit F2 for the inner ring, outer ring, rolling elements, and cage of the bearing is also not the same.

[0143] As a preferred embodiment, obtaining the second fault characteristic amplitude of the wheel from the spectral information of the second target vibration signal includes:

[0144] The maximum value of the current frequency domain signal in the spectrum information of the second target vibration signal is determined, and the maximum value of the current frequency domain signal is determined as the second fault characteristic amplitude of the wheel.

[0145] When detecting wheel faults, the maximum value of the current frequency domain signal in the spectrum information of the second target vibration signal is directly determined as the second fault characteristic amplitude of the wheel. That is, the maximum amplitude is found in the spectrum information of the second target vibration signal, which is the maximum value of the current frequency domain signal. In other words, when the wheel has a fault, it is represented by the position with the largest amplitude fluctuation in the spectrum information of the second target vibration signal.

[0146] As a preferred embodiment, determining whether the second fault characteristic amplitude of the wheel meets the wheel fault characteristic criteria includes:

[0147] Determine the target fault characteristic frequency corresponding to the second fault characteristic amplitude from the spectral information of the second target vibration signal;

[0148] Determine the rotational frequency of the wheels;

[0149] The effective order of polygonal faults in wheels is determined based on the ratio between the target fault characteristic frequency and the rotational frequency.

[0150] Obtain and determine the N second fault feature amplitudes that are continuously determined before the current time as N fault feature amplitudes to be measured;

[0151] Obtain and determine the N valid orders that are consecutively determined before the current time as the N valid orders to be tested;

[0152] If the amplitudes of all N test fault features are greater than the wheel alarm limit, and the N test valid orders are the same, then the second fault feature amplitude of the wheel is determined to meet the wheel fault feature standard.

[0153] In the field of rail transportation, wheel polygons are a common and serious wheel tread defect. It doesn't refer to a polygonal geometric shape, but rather to the periodic, uneven wear on the circumferential tread of a wheel after long-term operation, causing the wheel radius to exhibit alternating high and low frequency fluctuations of a fixed order along the circumference. Specifically, an ideal wheel is a perfect circle with a constant radius. However, in actual operation, due to complex interactions between the wheel and rail (such as stick-slip vibration, braking system characteristics, material fatigue, and track resonance), the wheel tread undergoes non-uniform wear or plastic deformation. This wear is not randomly distributed but forms peaks and troughs that repeat according to a specific pattern, and its waveform, when unfolded on the circumference, approximates a sine or cosine wave. When the wavelength of this waveform reaches a specific value, it is called an "N-order polygon," where N represents the number of peaks (or troughs) appearing within one circumference of the wheel.

[0154] When polygonal wheels roll, they generate periodic impacts on the rails, exciting high-frequency vibrations in the wheel-rail system and producing a piercing rolling noise that affects passenger comfort and the surrounding environment. Strong periodic impact loads significantly shorten the fatigue life of critical components such as wheels, rails, bearings, bogies, and even the car body structure, leading to a surge in maintenance costs. Severe high-order polygons may cause abnormally increased wheel-rail forces, potentially leading to derailment in extreme cases and posing a potential threat to train safety.

[0155] Once a wheel polygon is formed, it will persist for a considerable period. Therefore, accurate identification of the wheel polygon can be achieved by continuously judging the fault characteristics of the same wheel. After determining the amplitude of the second fault characteristic, the horizontal coordinate corresponding to the amplitude of the second fault characteristic, i.e., the target fault characteristic frequency, is determined from the spectral information of the second target vibration signal. The effective order Pt of the wheel polygon is determined by dividing the target fault characteristic frequency by the rotational frequency.

[0156] The order refers to the number of times the radius of a wheel changes periodically within one revolution, that is, the number of bulges (peaks) or troughs (valleys) that appear on the circumference. For example, a 20th-order polygon means that for every revolution of the wheel, there will be 20 periodic radius changes and corresponding impact excitations.

[0157] Different orders of polygons correspond to different fault mechanisms and excitation frequencies. By calculating the order, the frequency of the detected vibration signal can be precisely correlated with the rotational frequency of the wheel, thereby determining the dominant order of the polygon. This is far superior to simple diagnostics that only provide amplitude magnitude, and can provide clear guidance for maintenance, for example, indicating whether it is the 10th or 30th order of wear that needs to be addressed.

[0158] To ensure the effectiveness and accuracy of wheel condition detection, a number of consecutive tests N is set, where N is greater than or equal to 20. The amplitude of the second fault characteristic and the effective order Pt of the same wheel are statistically analyzed for N consecutive tests to determine the wheel condition.

[0159] The wheel alarm limit F4 is set as the fault characteristic standard for the wheel polygon. It is determined whether the amplitude of the second fault characteristic of the same wheel exceeds the wheel alarm limit F4 and whether the effective order of the N consecutive occurrences is the same. If the condition is met, the amplitude of the second fault characteristic of the wheel is determined to meet the wheel fault characteristic standard, and a wheel polygon alarm is reported. For example, if the amplitude of the second fault characteristic determined at the current moment is the 20th second fault characteristic amplitude, and the amplitudes of the second fault characteristic determined from the first to the 20th moment before this moment are 20 fault characteristic amplitudes to be tested, then the effective order determined at the current moment is the 20th effective order, and the effective orders determined from the first to the 20th moment before this moment are 20 effective orders to be tested. Only when the amplitudes of the 20 fault characteristics to be tested exceed the wheel alarm limit F4 and the 20 effective orders to be tested are the same, is the amplitude of the second fault characteristic of the wheel determined to meet the wheel fault characteristic standard, and a vehicle fault alarm is issued.

[0160] As a preferred embodiment, after obtaining and determining the N valid orders consecutively determined before the current time as the N valid orders to be measured, the method further includes:

[0161] If the amplitude values ​​of N fault characteristics to be tested are not all greater than the wheel alarm limit, but are all greater than the wheel warning limit, and the effective order of the N fault characteristics to be tested is the same, then a wheel fault warning is issued.

[0162] If the amplitude of all N test fault features is not greater than the wheel warning limit, and the N test valid orders are the same, then it is determined that there is no fault in the wheel.

[0163] If a wheel warning limit F3 is set for the wheel polygon feature index, and the amplitude of the second fault feature of the same wheel does not exceed the wheel alarm limit F4 for N consecutive times, or the effective order of the N consecutive tests is different, then to avoid false alarms, no vehicle fault alarm will be issued. Instead, it will be further determined whether the amplitude of the second fault feature of the same wheel for N consecutive times exceeds the wheel warning limit F3 and whether the effective order of the N consecutive tests is the same. If the conditions are met, a wheel polygon warning will be issued; otherwise, the wheel will be reported as normal. Based on this, an alarm can be issued in a timely manner when a wheel has a fault, and a warning can be issued when a wheel has a fault but it is not serious, so that staff can be informed of the wheel status in a timely manner.

[0164] Specifically, it can be determined whether the amplitude values ​​of N fault features to be tested are all greater than the wheel alarm limit and whether the N valid orders to be tested are the same. If so, a wheel fault alarm is triggered. If not, it can be determined whether the amplitude values ​​of N fault features to be tested are all greater than the wheel warning limit and whether the N valid orders to be tested are the same. If so, a wheel fault warning is triggered. If not, it is determined that there is no fault in the wheel.

[0165] The wheel alarm limit is greater than the wheel warning limit.

[0166] For example, among 20 measured fault feature amplitudes to be measured, only 15 measured fault feature amplitudes are greater than the wheel alarm limit value, and the other 5 measured fault feature amplitudes are not greater than the wheel alarm limit value but greater than the wheel warning limit value, and the 20 measured effective orders are the same. Then, instead of giving a wheel fault alarm, a vehicle fault warning is given. If among the 20 measured fault feature amplitudes to be measured, 5 measured fault feature amplitudes are greater than the wheel alarm limit value, 5 measured fault feature amplitudes are not greater than the wheel alarm limit value but greater than the wheel warning limit value, and 10 measured fault feature amplitudes are not greater than the wheel warning limit value, and the 20 measured effective orders are the same, it is determined that there is no fault in the wheel.

[0167] As a preferred embodiment, determining the effective order of the polygonal fault of the wheel based on the ratio between the target fault feature frequency and the rotational frequency includes:

[0168] Determining the ratio between the target fault feature frequency and the rotational frequency as the calculated order of the polygonal fault of the wheel;

[0169] Performing a rounding operation on the calculated order to determine the theoretical order;

[0170] Determining the order deviation based on the absolute value of the difference between the calculated order and the theoretical order;

[0171] If the order deviation is less than the deviation limit value, determining the theoretical order as the effective order;

[0172] If the order deviation is not less than the deviation limit value, updating the maximum value of the frequency domain signals other than the maximum value of the current frequency domain signal in the frequency spectrum information of the second target vibration signal to the maximum value of the current frequency domain signal, and returning to the step of determining the second fault feature amplitude of the wheel by taking the maximum value of the current frequency domain signal.

[0173] Dividing the target fault feature frequency by the rotational frequency to obtain the calculated order Pc1 of the wheel polygon. According to the formation principle of the wheel polygon, the theoretical order of the wheel polygon is an integer. Performing a rounding operation on the above calculated order Pc1 can obtain the theoretical order Pt1 of the wheel polygon. Calculating the absolute value of the difference between the calculated order Pc1 and the theoretical order Pt1 of the wheel polygon as the order deviation Pq1, that is, Pq1 = |Pc1 - Pt1|. Setting the deviation limit value Pq, Pq can be but not limited to less than 0.2. When Pq1 < Pq, the theoretical order can be used as the effective order, that is, the effective order is Pt1. Otherwise, extracting the second largest value in the frequency spectrum information of the second target vibration signal as the second fault feature amplitude of the wheel polygon, dividing the corresponding target fault feature frequency by the rotational frequency to obtain the calculated order Pc2 of the wheel polygon, and so on, until the effective order of the wheel polygon is calculated as Pt. Based on this, the effective order of the wheel polygon can be determined more accurately, and the wheel can be detected more accurately.

[0174] It should be further explained that updating the maximum value of the frequency domain signal other than the current maximum value in the spectrum information of the second target vibration signal to the current maximum value means updating the maximum value of the frequency domain signal among the frequency domain signals that are smaller than the current maximum value in the spectrum information of the second target vibration signal to the current maximum value. If the first and second largest values ​​in the entire spectrum information of the second target vibration signal cannot determine the effective order, then the effective order is calculated based on the third largest value.

[0175] As a preferred embodiment, obtaining the third fault characteristic amplitude of the rail from the spectral information of the third target vibration signal includes:

[0176] The maximum value and effective value of the rail frequency domain signal are obtained from the spectral information of the vibration signal of the third target.

[0177] When detecting rail faults, the maximum amplitude value, i.e., the maximum value of the rail frequency domain signal Qmax, is extracted from the spectral information of the vibration signal of the third target. At the same time, the effective value of the rail frequency domain signal Qrms is extracted from the spectral information of the vibration signal of the third target. When a rail fault occurs, the corresponding fault characteristics are the maximum value of the rail frequency domain signal Qmax and the effective value of the rail frequency domain signal Qrms. Therefore, rail fault detection can be performed based on the maximum value of the rail frequency domain signal Qmax and the effective value of the rail frequency domain signal Qrms.

[0178] As a preferred embodiment, determining whether the amplitude of the third fault characteristic of the rail meets the fault characteristic standard of the rail includes:

[0179] The maximum value of the rail frequency domain signal and the average value of the rail frequency domain signal are determined as rail fault characteristic indicators.

[0180] Determine the total time interval during which all carriages of the vehicle pass the target test point on the rail.

[0181] Obtain the rail fault characteristic indicators of each car in the vehicle within the total time period.

[0182] The maximum value of the rail fault characteristic index for each car is determined from the rail fault characteristic indexes of each car within the total time period.

[0183] If the maximum value of the rail fault characteristic index corresponding to each carriage is greater than the rail fault alarm limit, then the third fault characteristic amplitude of the rail is determined to meet the rail fault characteristic standard.

[0184] Rail corrugation is a typical periodic surface damage phenomenon in rail transit systems. It manifests as periodic, wavy, uneven wear along the longitudinal direction of the rail top surface or gauge angle, with its waveform exhibiting a regular, approximately sinusoidal or cosine-like undulation along the track extension direction. The formation mechanism of this phenomenon is complex, resulting from self-excited vibration or resonance generated in the wheel-rail system under specific conditions. When a vehicle runs at a specific speed, the interaction forces between the wheel and rail (such as the stick-slip effect) excite vibrations at specific frequencies. This vibration is fed back to the contact interface, causing periodic plastic deformation or wear of the rail material. Over a long period, this eventually forms corrugation with a fixed wavelength. Its wavelength range typically varies from tens to hundreds of millimeters and is closely related to the excitation frequency and train speed.

[0185] Based on this, a weighted average of the maximum value Qmax and the effective value Qrms of the rail frequency domain signal is used to calculate the rail fault characteristic index Q = (Qmax + Qrms) / 2. Simultaneously, the rail corrugation wavelength w = vehicle speed / the characteristic frequency corresponding to the maximum value of the rail frequency domain signal can be calculated, allowing staff to identify the rail condition. When the vehicle travels on the rail, for a target test point on the rail, the front of the vehicle passes the target test point first, and then the rear of the vehicle passes the target test point. The total time from the front to the rear of the vehicle passing the target test point is the total time period. This total time period includes multiple sampling points; that is, each car will have multiple rail fault characteristic indices within the total time period. To determine whether a fault exists on the rail, the maximum value of each rail fault characteristic index for each car within the total time period is set as the maximum value of the rail fault characteristic index. If the maximum value of the rail fault characteristic index corresponding to each car is greater than the rail fault alarm limit, then the third fault characteristic amplitude of the rail is determined to meet the rail fault characteristic standard, meaning that a fault exists at the target test point of the rail.

[0186] For example, a vehicle consists of four carriages, with a total length of L and a speed of V. The total time interval t from the front of the vehicle passing the target measurement point to the rear of the vehicle passing the target measurement point is t = L / V. Each carriage contains the aforementioned vibration signals, including vibration signals from the left and right sides. Within the total time interval t, there are multiple sampling points. Therefore, each carriage corresponds to the spectral information of a third target vibration signal from the left and a third target vibration signal from the right at each sampling point. For the first carriage, the maximum value Qmax and the effective value Qrms of the rail frequency domain signal in the spectral information of the third target vibration signal from the left at the first sampling point within the total time interval t are determined, and the first rail fault characteristic index Q is calculated. The maximum value Qmax and the effective value Qrms of the rail frequency domain signal in the spectral information of the third target vibration signal from the left at the second sampling point within the total time interval t are also determined. The value Qrms is calculated, and the second rail fault characteristic index Q is calculated until the maximum value Qmax and the effective value Qrms of the rail frequency domain signal in the spectrum information of the third target vibration signal on the left side of the last sampling point of the first car within the total time period t are determined. The last rail fault characteristic index Q is calculated, and the largest rail fault characteristic index Q is determined from the first rail fault characteristic index Q to the last rail fault characteristic index Q as the maximum value of the rail fault characteristic index on the left side of the first car. Based on this, the maximum values ​​of the left rail fault characteristic index on the left side of the second car to the fourth car are determined sequentially. Thus, based on the maximum values ​​of the left rail fault characteristic index on the left side of the first car to the fourth car, it is determined whether rail corrugation has occurred at the target test point on the left side. That is, if the maximum values ​​of the left rail fault characteristic index on the left side of each car are all greater than the rail fault alarm limit, then the third fault characteristic amplitude of the rail is determined to meet the rail fault characteristic standard. For the fault detection of the target test point on the right side of the rail, please refer to the above-described fault detection process for the target test point on the left side. This application does not limit this process.

[0187] As a preferred embodiment, after determining the maximum value of the rail fault characteristic index corresponding to each car from the various rail fault characteristic indices of each car within the total time period, the method further includes:

[0188] If the maximum values ​​of the rail fault characteristic indicators corresponding to each car are not all greater than the rail fault alarm limit, but are all greater than the rail fault warning limit, then a rail fault warning will be issued.

[0189] If the maximum value of the rail fault characteristic index corresponding to each car is 28 and does not exceed the rail fault warning limit, then it is determined that there is no rail fault.

[0190] Set a rail fault warning limit F5 and a rail fault alarm limit F6, where the maximum value of the rail fault characteristic index is greater than the rail fault warning limit F5. First, determine whether the maximum value of the rail fault characteristic index of multiple cars corresponding to the same side rail exceeds the rail fault alarm limit F6. If the condition is met, report a rail corrugation alarm for that side. If the condition is not met, determine whether the maximum value of the rail fault characteristic index of multiple cars corresponding to the same side rail exceeds the rail fault warning limit F5. If the condition is met, report a rail corrugation warning for that side; otherwise, report that the rail is normal.

[0191] Specifically, when determining whether a fault has occurred on the left rail, the maximum value of the fault characteristic index of the left rail of each car is compared with the rail fault alarm limit F6. If the maximum value of the fault characteristic index of the left rail of each car is greater than the rail fault alarm limit F6, then the target test point on the corresponding left rail has a fault and an alarm is triggered. If the maximum value of the fault characteristic index of the left rail of each car is not greater than the rail fault alarm limit F6, but is greater than the rail fault warning limit F5, then a fault warning is issued for the left rail. If the maximum value of the fault characteristic index of the left rail of each car is not greater than the rail fault warning limit F5, then the left rail can be determined to be normal. When determining whether a fault has occurred on the right-side rail, the maximum value of the fault characteristic index of the right-side rail of each car is compared with the rail fault alarm limit F6. If the maximum value of the fault characteristic index of the right-side rail of each car is greater than the rail fault alarm limit F6, then the target test point on the corresponding right-side rail has a fault and an alarm is triggered. If the maximum value of the fault characteristic index of the right-side rail of each car is not greater than the rail fault alarm limit F6, but is greater than the rail fault warning limit F5, then a right-side rail fault warning is triggered. If the maximum value of the fault characteristic index of the right-side rail of each car is not greater than the rail fault warning limit F5, then the right-side rail can be determined to be normal.

[0192] It should be noted that both the left and right sides of the rail include multiple target test points, which may be continuous or discontinuous. This application does not limit this, and the test points may be the pre-determined target test points on the rail with the highest probability of rail corrugation.

[0193] It should also be noted that this application also discloses a rail transit vehicle bearing fault detection system, which mainly includes a data acquisition module, a data transmission module, a data analysis module, and a human-machine interface module.

[0194] The system comprises several modules: a data acquisition module, a data processing module, and a data analysis module. The data acquisition module includes sensors for collecting vibration and velocity signals from the bearing under test, along with corresponding preprocessors. The data transmission module includes a vehicle network system for transmitting the vibration and velocity signals to the data analysis module. The data analysis module contains vehicle-level and train-level data processors for analyzing and processing the vibration and velocity signals to detect bearing faults. It also includes vehicle-level and train-level data storage devices for storing vehicle-level and train-level vibration and velocity signals, respectively. The human-machine interface module primarily includes a results display to show the analysis results from the data analysis module and provide guidance to train operators.

[0195] The data acquisition module is used to collect and preprocess vibration data of the axle box and speed data of the vehicle. The raw vibration and speed signals are cleaned and converted from analog to digital by the preprocessor, and the digital signals are transmitted to the vehicle-level data processor or the train-level data processor.

[0196] The data transmission module is used to transmit the vibration and speed signals collected by the sensors to the data analysis module and the data storage. Data transmission can be achieved using the existing train control network or by establishing a dedicated fault diagnosis network. Data transmission can be wired (e.g., RS232 or Ethernet) or wireless (e.g., Wi-Fi), and this application does not limit the specific method used.

[0197] The data analysis module is used to analyze and process vibration and velocity signals to detect and identify faults in the bearing under test.

[0198] The data processing flow of the data analysis module follows the method embodiment described above.

[0199] The human-machine interface (HMI) module includes a results display screen to show the analysis results from the data analysis module, providing guidance to train operators. Through an interactive interface, the HMI module visualizes the bearing fault detection results and allows for detailed parameter viewing and historical parameter review as needed.

[0200] Please refer to Figure 2 , Figure 2 This application provides a structural schematic diagram of a fault detection system for a rail transit system, which includes:

[0201] The first acquisition unit 21 is used to acquire the vibration signal at the axle box of the vehicle and filter the vibration signal for the target frequency band corresponding to each test position of the rail transit system to determine the target vibration signal of the target frequency band corresponding to each test position.

[0202] Analysis unit 22 is used to perform spectrum analysis on the target vibration signal corresponding to each test position to obtain the spectrum information of each target vibration signal;

[0203] The second acquisition unit 23 is used to acquire the fault characteristic amplitude of each test location from the spectrum information of each target vibration signal;

[0204] The first determining unit 24 is used to determine whether the fault characteristic amplitude of each test location meets the corresponding fault characteristic standard.

[0205] The second determining unit 25 is used to determine that there is a fault at the corresponding test location if the fault characteristic amplitude meets the fault characteristic standard of the corresponding test location.

[0206] For an introduction to the fault detection system for rail transit systems provided by this invention, please refer to the above method embodiments; further details of this invention will not be repeated here.

[0207] Please refer to Figure 3 , Figure 3 This application provides a structural schematic diagram of a fault detection device for a rail transit system, which includes:

[0208] Memory 31 is used to store computer programs;

[0209] The processor 32 is used to implement the steps of the fault detection method for rail transit systems as described above when executing a computer program.

[0210] For a description of the fault detection device for rail transit systems provided by the present invention, please refer to the above method embodiments; the present invention will not be described again here.

[0211] Please refer to Figure 4 , Figure 4 This is a schematic diagram of the structure of a computer-readable storage medium provided in this application. The computer-readable storage medium 41 stores a computer program 42. When the computer program 42 is executed by the processor 32, it implements the steps of the road traffic system fault detection method described above.

[0212] For a description of the computer-readable storage medium provided by the present invention, please refer to the above method embodiments; the present invention will not be described again here.

[0213] It should also be noted that, in this specification, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0214] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A fault detection method for a rail transit system, characterized in that, include: The vibration signal at the axle box of the vehicle is acquired, and the vibration signal is filtered for the target frequency band corresponding to each test location in the rail transit system to determine the target vibration signal of the target frequency band corresponding to each test location. Spectral analysis is performed on the target vibration signal corresponding to each of the measured positions to obtain the spectral information of each target vibration signal; The fault characteristic amplitude of each of the test locations is obtained from the spectral information of each of the target vibration signals; Determine whether the fault characteristic amplitude of each of the locations to be tested conforms to the corresponding fault characteristic standard; If the fault characteristic amplitude meets the fault characteristic standard of the corresponding test location, then it is determined that there is a fault at the corresponding test location.

2. The fault detection method for rail transit systems as described in claim 1, characterized in that, The locations to be tested include bearings, wheels, and rails; The vibration signal is filtered according to the target frequency band corresponding to each test location in the rail transit system to determine the target vibration signal of each test location in the target frequency band, including: Determine the first minimum fault modulation frequency when the bearing has a fault, the first maximum fault modulation frequency when the wheel has a fault, and the second minimum fault modulation frequency and the second maximum fault modulation frequency when the rail has a fault. The vibration signal is high-pass filtered based on the first lowest fault modulation frequency to determine the first target vibration signal corresponding to the bearing with a frequency higher than the first lowest fault modulation frequency. The vibration signal is low-pass filtered based on the first highest fault modulation frequency to determine the second target vibration signal corresponding to the wheel with a frequency lower than the first highest fault modulation frequency. The vibration signal is bandpass filtered based on the second lowest fault modulation frequency and the second highest fault modulation frequency to determine a third target vibration signal whose frequency corresponds to the rail in the target frequency band between the second lowest fault modulation frequency and the second highest fault modulation frequency.

3. The fault detection method for rail transit systems as described in claim 2, characterized in that, The fault characteristic amplitude of each of the measured locations is obtained from the spectral information of each of the target vibration signals, including: The first fault characteristic amplitude of the bearing is obtained from the spectral information of the first target vibration signal; The second fault characteristic amplitude of the wheel is obtained from the spectral information of the second target vibration signal; The third fault characteristic amplitude of the rail is obtained from the spectral information of the third target vibration signal; Determine whether the fault characteristic amplitude at each of the locations to be tested conforms to the corresponding fault characteristic standard, including: Determine whether the amplitude of the first fault characteristic of the bearing meets the fault characteristic standard of the bearing; Determine whether the second fault characteristic amplitude of the wheel meets the fault characteristic standard of the wheel; Determine whether the amplitude of the third fault characteristic of the rail meets the fault characteristic standard of the rail.

4. The fault detection method for rail transit systems as described in claim 3, characterized in that, Before obtaining the first fault characteristic amplitude of the bearing from the spectral information of the first target vibration signal, the method further includes: The fault characteristic frequencies of each component under test of the bearing are determined based on the bearing parameters and the bearing's operating conditions. Obtaining the first fault characteristic amplitude of the bearing from the spectral information of the first target vibration signal includes: The first fault characteristic amplitude values ​​of the fault characteristic frequencies of each component under test of the bearing, from the first target vibration signal spectrum information, are obtained from the first harmonic to the Mth harmonic of the fault characteristic frequency.

5. The fault detection method for rail transit systems as described in claim 4, characterized in that, Determining whether the amplitude of the first fault characteristic of the bearing meets the fault characteristic criteria of the bearing includes: Calculate the average fault characteristic amplitude of the first fault characteristic amplitude from the first harmonic to the Mth harmonic of the fault characteristic frequency of each component under test of the bearing; Determine the alarm limit values ​​corresponding to each component under test of the bearing; If the average fault characteristic amplitude is greater than the alarm limit of the corresponding component under test, then the first fault characteristic amplitude of the bearing is determined to meet the fault characteristic standard of the bearing.

6. The fault detection method for rail transit systems as described in claim 5, characterized in that, After calculating the average fault characteristic amplitude of the first fault characteristic amplitude from the first harmonic to the Mth harmonic of the fault characteristic frequency of each component under test of the bearing, the method further includes: Determine the warning limit for each component under test of the bearing; the alarm limit is greater than the warning limit. If the average fault characteristic amplitude is greater than the warning limit of the corresponding component under test, but not greater than the alarm limit of the corresponding component under test, then a fault warning is issued for the corresponding component under test. If the average fault characteristic amplitude is not greater than the warning limit of the corresponding component under test, then it is determined that the corresponding component under test does not have a fault.

7. The fault detection method for rail transit systems according to claim 3, characterized in that, Obtaining the second fault characteristic amplitude of the wheel from the spectral information of the second target vibration signal includes: The maximum value of the current frequency domain signal in the spectral information of the second target vibration signal is determined, and the maximum value of the current frequency domain signal is determined as the second fault characteristic amplitude of the wheel.

8. The fault detection method for a rail transit system as described in claim 7, characterized in that, Determining whether the second fault characteristic amplitude of the wheel meets the fault characteristic standard of the wheel includes: Determine the target fault characteristic frequency corresponding to the second fault characteristic amplitude from the spectral information of the second target vibration signal; Determine the rotational frequency of the wheel; The effective order of the polygonal fault of the wheel is determined based on the ratio between the target fault characteristic frequency and the rotational frequency. Obtain and determine the N second fault feature amplitudes that are continuously determined before the current time as N fault feature amplitudes to be measured; Obtain and determine the N valid orders that are consecutively determined before the current time as the N valid orders to be tested; If the amplitude values ​​of all N fault features to be tested are greater than the wheel alarm limit, and the N valid orders to be tested are the same, then the second fault feature amplitude value of the wheel is determined to meet the fault feature standard of the wheel.

9. The fault detection method for a rail transit system as described in claim 8, characterized in that, After obtaining and determining the N consecutive valid orders determined before the current time as the N valid orders to be tested, the process also includes: If the amplitude values ​​of the N fault features to be tested are not all greater than the wheel alarm limit, but are all greater than the wheel warning limit, and the N valid orders to be tested are the same, then a wheel fault warning is issued. If the amplitude of all N test fault features is not greater than the wheel warning limit, and the effective order of the N test features is the same, then it is determined that the wheel does not have a fault.

10. The fault detection method for a rail transit system as described in claim 8, characterized in that, Determining the effective order of the polygonal fault of the wheel based on the ratio between the target fault characteristic frequency and the rotational frequency includes: The ratio between the target fault characteristic frequency and the rotational frequency is determined as the calculation order of the polygonal fault of the wheel; The calculated order is rounded down to determine the theoretical order; The order deviation is determined based on the absolute value of the difference between the calculated order and the theoretical order; If the order deviation is less than the deviation limit, then the theoretical order is determined as the effective order; If the order deviation is not less than the deviation limit, then the maximum value of the frequency domain signal other than the current maximum value of the frequency domain signal in the spectrum information of the second target vibration signal is updated to the current maximum value of the frequency domain signal, and the step of determining the current maximum value of the frequency domain signal as the second fault characteristic amplitude of the wheel is returned.

11. The fault detection method for a rail transit system according to claim 3, characterized in that, The third fault characteristic amplitude of the rail is obtained from the spectral information of the third target vibration signal, including: The maximum value and effective value of the rail frequency domain signal are obtained from the spectral information of the vibration signal of the third target.

12. The fault detection method for a rail transit system according to claim 11, characterized in that, Determining whether the third fault characteristic amplitude of the rail meets the fault characteristic standard of the rail includes: The maximum value of the rail frequency domain signal and the average value of the rail frequency domain signal are determined as rail fault characteristic indicators. Determine the total time period during which all carriages of the vehicle pass the target test point on the rail; Obtain the rail fault characteristic indicators of each car of the vehicle within the total time period. The maximum value of the rail fault characteristic index corresponding to each of the various rail fault characteristic indices of each of the various rail fault characteristic indices within the total time period is determined from each of the rail fault characteristic indices of each of the various ... indices. If the maximum value of the rail fault characteristic index corresponding to each of the carriages is greater than the rail fault alarm limit, then the third fault characteristic amplitude of the rail is determined to meet the rail fault characteristic standard.

13. The fault detection method for a rail transit system according to claim 12, characterized in that, After determining the maximum value of the rail fault characteristic index corresponding to each of the various rail fault characteristic indices for each of the various rail fault characteristic indices within the total time period, the method further includes: If the maximum value of the rail fault characteristic index corresponding to each of the carriages is not greater than the rail fault alarm limit, but is greater than the rail fault warning limit, then a rail fault warning is issued. If the maximum value of the rail fault characteristic index corresponding to each of the carriages is not greater than the rail fault warning limit, then it is determined that the rail has no fault.

14. A fault detection device for a rail transit system, characterized in that, include: Memory, used to store computer programs; A processor, configured to, when executing a computer program, implement the steps of the rail transit system fault detection method as described in any one of claims 1-13.

Images