CN113567128B - Method, equipment and storage medium for precisely extracting and diagnosing fault characteristics of train bearing - Google Patents
Method, equipment and storage medium for precisely extracting and diagnosing fault characteristics of train bearing Download PDFInfo
- Publication number
- CN113567128B CN113567128B CN202110846485.6A CN202110846485A CN113567128B CN 113567128 B CN113567128 B CN 113567128B CN 202110846485 A CN202110846485 A CN 202110846485A CN 113567128 B CN113567128 B CN 113567128B
- Authority
- CN
- China
- Prior art keywords
- signal
- frequency
- signal segment
- kurtosis
- segment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/04—Bearings
- G01M13/045—Acoustic or vibration analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
- G06F17/148—Wavelet transforms
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Abstract
The invention discloses a method, equipment and a storage medium for precisely extracting and diagnosing fault characteristics of a train bearing, wherein the method comprises the steps of layering and segmenting a vibration signal of a bearing to be detected of a train; processing each signal segment based on a signal separation operator SSO to obtain the kurtosis of each signal segment; determining a signal segment with the highest kurtosis in all signal segments, and constructing a band-pass filter according to the center frequency and the frequency resolution of the signal segment with the highest kurtosis; filtering the vibration signal by using the band-pass filter; and obtaining fault characteristic frequency based on the filtered vibration signal, and identifying a fault state. According to the invention, the kurtosis calculation is carried out by adopting the signal separation operator SSO, a plurality of instantaneous frequencies can be extracted at a certain moment, very similar frequency components can be separated, more accurate characteristic frequency can be obtained, and the phenomena of misdiagnosis and missed diagnosis of faults are avoided.
Description
Technical Field
The invention belongs to the technical field of train bearing fault diagnosis, and particularly relates to a method, equipment and a storage medium for precisely extracting and diagnosing train bearing fault characteristics.
Background
In a high-speed train bogie, axle box bearings are key components of a vehicle running part and play a key role in ensuring the driving safety. Because the bearing is in a severe working environment for a long time, faults such as burning, corrosion, pockmarks, rolling skin, dents, cracks, bruises, scratches and the like on a rolling working surface of the bearing are easy to generate; the degree of the fault can easily reach or even exceed the class c standard issued by the national ministry of railways, and the potential threat to the running safety of railways is higher in the precision of rolling bearings compared with general mechanical parts, so that the use is also careful and careful accordingly. Once axle box bearings fail, the whole train cannot run, transportation lines are blocked, and a large amount of passenger flow and cargo flow is reserved. The negative effects are even greater especially when the passenger train malfunctions. Since the axle box bearings are mounted inside the axle boxes, it is also a very difficult task to make an emergency repair of the failed axle box bearings in the vehicle section. Therefore, the method can prevent and reduce the occurrence of axle box bearing faults, realize the precise diagnosis of the axle box bearing early faults, and is very important for the safe operation of high-speed motor train unit vehicles. The axle box bearing of the high-speed motor train unit is a double-row tapered roller bearing and consists of an inner ring, an outer ring, a rolling body and a retainer, wherein the inner ring of the bearing and a rotor run at a high speed to bear a large load, so that the bearing is easy to break down. Vibration signals and noise generated during operation of a vehicle due to various causes in a working environment are complicated. Failure or changes in bearing compliance are the two leading causes of bearing vibration and noise. When the bearing of the gearbox fails, the amplitude of the vibration signal of the bearing is easily increased. A local failure of one element in a bearing, when interacting with another normal element, also causes abrupt changes in surface contact stresses, resulting in a vibration excitation of extremely short duration.
At present, axle box temperature sensors are installed on high-speed motor train units in China, the axle box temperature sensors on the motor train units can monitor the temperature of bearings in real time, but alarm is only carried out when the temperature rise exceeds a threshold value, and false alarm and missed alarm are easily caused due to the influence of service environment. The bearing vibration signal and the bearing early failure or potential health hidden danger keep synchronism and sensitivity, so that the bearing online detection theory and technology based on the vibration signal are rapidly developed and widely applied. In the prior art, common fault feature extraction methods such as statistical parameters, wavelet transformation, wignerviller distribution and the like have respective defects, and if the fault feature extraction methods are applied to actual engineering, the diagnosis result is unstable, so that the phenomena of misdiagnosis and missed diagnosis of faults are generated; secondly, most of the existing research methods do not consider a plurality of factors in the running process of the high-speed train, such as influence of unsmooth lines, load change, speed change, wheel pair faults and the like on axle box bearing vibration signals, so that the axle box bearing fault characteristics are hidden in the axle box vibration signals of the high-speed motor train unit, and the online monitoring effect of the axle box bearing of the high-speed motor train unit is not ideal.
Disclosure of Invention
In order to solve the problems that the existing methods for precisely extracting and diagnosing the fault characteristics of the train bearing adopt statistical parameters, wavelet transformation, wingerwire distribution and the like, the diagnosis result is unstable, and the phenomena of misdiagnosis and missed diagnosis of the fault exist, the invention provides the method, the equipment and the storage medium for precisely extracting and diagnosing the fault characteristics of the train bearing, wherein a signal separation operator SSO is adopted for carrying out kurtosis calculation, a plurality of instantaneous frequencies can be extracted at a certain moment, very similar frequency components can be separated, more precise characteristic frequencies can be obtained, and the phenomena of misdiagnosis and missed diagnosis of the fault are avoided.
The invention is realized by the following technical scheme:
the invention provides a train bearing fault feature precise extraction and diagnosis method in a first aspect, which comprises the following steps:
acquiring a vibration signal of a bearing to be detected of a train;
carrying out layered segmentation on the vibration signal to obtain at least one signal segment;
processing each signal segment in the at least one signal segment based on a signal separation operator SSO to obtain the kurtosis of each signal segment;
determining a signal segment with the highest kurtosis in all signal segments, and constructing a band-pass filter according to the center frequency and the frequency resolution of the signal segment with the highest kurtosis;
filtering the vibration signal by using the band-pass filter;
and obtaining fault characteristic frequency based on the filtered vibration signal, and identifying a fault state.
Vibration signals of the bearings contain a large amount of noise in the running process of the high-speed motor train unit, and the vibration characteristics of early faults of the bearings are hidden in strong background noise. According to the scheme, the vibration signals are segmented in a layered mode, the kurtosis of each signal segment is calculated through the signal separation operator SSO, and compared with the existing methods of statistical parameters, wavelet transformation, wingeville distribution and the like, only one instantaneous frequency can be obtained at a certain moment, a plurality of instantaneous frequencies can be extracted at a certain moment, namely very similar frequency components can be separated, more accurate characteristic frequencies can also be obtained, the method has great significance for accurately extracting the characteristic frequencies under the condition that the bearing fault characteristic frequencies shift due to various factors in the running process of the high-speed motor train unit, and the accuracy of on-line monitoring of the axle box bearing of the high-speed motor train unit is improved. A band-pass filter is constructed according to the center frequency and the frequency resolution of the signal segment with the highest kurtosis by determining the signal segment with the highest kurtosis, and weak fault characteristics hidden in noise can be effectively extracted after band-pass filtering.
In one possible design, the processing each of the at least one signal segment based on the signal separation operator SSO to obtain a kurtosis of each signal segment includes:
performing signal separation operator SSO calculation on each signal segment in the at least one signal segment to obtain a frequency set and an amplitude set of each signal segment in each time;
determining the instantaneous frequency corresponding to each time of each signal segment according to the frequency set and the amplitude set of each time of each signal segment;
calculating the absolute value of the SSO output result to obtain the instantaneous amplitude of the instantaneous frequency;
and calculating the complex envelope of the instantaneous frequency and the instantaneous amplitude of each signal segment at all times to obtain the kurtosis of each signal segment.
In one possible design, the determining the instantaneous frequency for each signal segment at each time based on the set of frequencies and the set of amplitudes for each signal segment at each time includes:
clustering the frequency sets in each time after thresholding to obtain at least one disjoint frequency cluster in each time;
and determining the frequency corresponding to the maximum amplitude in each frequency cluster of at least one frequency cluster at each time, and taking the frequency as the instantaneous frequency at the moment.
In one possible design, the window width in the signal separation operator SSO is greater than or equal to 2 7 。
The window width in the signal separation operator SSO is related to the accuracy of the output of the signal separation operator SSO, which should be sufficiently large. However, the method of calculating kurtosis needs to be adapted to non-stationary processes, which means that the window width cannot be too long. According to the scheme, the window width in the signal separation operator SSO is set in the interval, so that the output precision of the signal separation operator SSO is met, and the calculation precision of the kurtosis is also met.
In one possible design, the determining a signal segment with a highest kurtosis among all signal segments includes:
establishing a frequency matching structure chart according to the kurtosis of each signal segment and the corresponding center frequency, or establishing a frequency resolution matching structure chart according to the kurtosis of each signal segment and the corresponding frequency resolution;
and determining the signal segment with the highest kurtosis in all the signal segments according to the frequency matching structure chart or the frequency resolution matching structure chart.
According to the scheme, the frequency matching structure chart or the frequency resolution matching structure chart is established, the resonant frequency band of the perimeter signal is accurately positioned, weak fault characteristics hidden in noise can be effectively extracted after band-pass filtering, and the comprehensiveness and the accuracy of bearing fault diagnosis are improved.
In one possible design, the hierarchically segmenting the vibration signal to obtain at least one signal segment includes:
determining the maximum allowable layering number according to the signal length of the vibration signal;
determining the actual layering number according to the signal length and the maximum allowable layering number of the vibration signal;
layering the vibration signals according to the actual layering number to obtain at least one layer of signals;
and segmenting each layer of signal in the at least one layer of signal to obtain at least one signal segment.
The invention provides a train bearing fault feature precision extraction and diagnosis device, which comprises a signal acquisition unit, a signal layering and segmenting unit, a kurtosis calculation unit, a band-pass filter construction unit and a fault state identification unit which are connected in sequence through signals,
the signal acquisition unit is used for acquiring a vibration signal of a bearing to be detected of the train;
the signal layering and segmenting unit is used for layering and segmenting the vibration signal to obtain at least one signal segment;
the kurtosis calculating unit is used for processing each signal segment in the at least one signal segment based on a signal separation operator SSO to obtain the kurtosis of each signal segment;
the band-pass filter constructing unit is used for determining a signal segment with the highest kurtosis in all signal segments, constructing a band-pass filter according to the center frequency and the frequency resolution of the signal segment with the highest kurtosis, and filtering the vibration signal by using the band-pass filter;
the fault state identification unit is used for obtaining fault characteristic frequency based on the filtered vibration signals and identifying fault states.
The invention provides a train bearing fault feature precise extraction and diagnosis device, which comprises a memory and a controller which are sequentially connected in a communication manner, wherein the memory is stored with a computer program, and the processor is used for reading the computer program and executing the train bearing fault feature precise extraction and diagnosis method in the first aspect and any one possibility.
A fourth aspect of the present invention provides a computer-readable storage medium, which stores instructions that, when executed on a computer, perform the method for precisely extracting and diagnosing train bearing fault characteristics according to the first aspect and any one of the possibilities.
Compared with the prior art, the invention at least has the following advantages and beneficial effects:
1. compared with the existing methods of statistical parameters, wavelet transformation, wingeville distribution and the like, the method can only obtain one instantaneous frequency at a certain moment, can extract a plurality of instantaneous frequencies at a certain moment, can separate very similar frequency components and can also obtain more accurate characteristic frequency, has great significance for precisely extracting the characteristic frequency under the condition that the characteristic frequency of the bearing fault is deviated due to various factors in the running process of the high-speed motor train unit, improves the accuracy of on-line monitoring of the high-speed motor train unit axle box bearing, and avoids misdiagnosis and missed diagnosis of the fault.
2. According to the method, the signal segment with the highest kurtosis is determined, the band-pass filter is constructed according to the center frequency and the frequency resolution of the signal segment with the highest kurtosis, and weak fault characteristics hidden in noise can be effectively extracted after band-pass filtering, so that the accuracy of train bearing fault diagnosis is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flow chart of a method for precisely extracting and diagnosing fault characteristics of a train bearing according to the present invention;
FIG. 2 is a flow chart of a method for obtaining kurtosis based on the signal separation operator SSO according to the present invention;
in fig. 3, a is an original signal x (t) obtained, B is a corresponding frequency spectrum, and C is an envelope spectrum of the original signal;
FIG. 4 is a diagram of a frequency matching architecture constructed in accordance with an embodiment of the present invention;
fig. 5 is a diagram of an analysis of a complex envelope of a filtered signal in accordance with an embodiment of the present invention.
Fig. 6 is a schematic block diagram of a train bearing fault diagnosis apparatus of the present invention.
Detailed Description
The invention is further described with reference to the following figures and specific embodiments. It should be noted that the description of the embodiments is provided to help understanding of the present invention, but the present invention is not limited thereto. Specific structural and functional details disclosed herein are merely illustrative of example embodiments of the invention. This invention may, however, be embodied in many alternate forms and should not be construed as limited to the embodiments set forth herein.
It should be understood that, for the term "and/or" as may appear herein, it is merely an associative relationship that describes an associated object, meaning that three relationships may exist, e.g., a and/or B may mean: a exists independently, B exists independently, and A and B exist simultaneously; for the term "/and" as may appear herein, which describes another associative object relationship, it means that there may be two relationships, e.g., a/and B, which may mean: a exists independently, and A and B exist independently; in addition, for the character "/" that may appear herein, it generally means that the former and latter associated objects are in an "or" relationship.
It should be understood that specific details are provided in the following description to facilitate a thorough understanding of example embodiments. However, it will be understood by those of ordinary skill in the art that the example embodiments may be practiced without these specific details. For example, systems may be shown in block diagrams in order not to obscure the examples in unnecessary detail. In other instances, well-known processes, structures and techniques may not be shown in unnecessary detail to avoid obscuring the examples.
As shown in fig. 1, a first aspect of the present invention discloses a method for precisely extracting and diagnosing train bearing fault features, which may be, but is not limited to, performed by a fault diagnosis device, where the fault diagnosis device may be software, or a combination of software and hardware, and the fault diagnosis device may be integrated in an intelligent device such as an intelligent mobile terminal, a tablet, a computer, etc. Specifically, the fault diagnosis method includes the following steps S001 to S006.
And S001, acquiring a vibration signal of the bearing to be detected of the train.
In the step, a vibration sensor is fixedly arranged on a bearing to be detected of the train to acquire a vibration signal of the bearing, the vibration sensor can be a vibration acceleration sensor, high-frequency sampling is carried out, and the measured information can be converted into information in other required forms according to a certain rule to be output. Intercepting a section of the output signal of the vibration accelerator sensor as a vibration signal x (t) of a bearing to be detected of the train, and taking the vibration signal x (t) as an original signal. Take the original signal in fig. 3 as an example.
And S002, carrying out layered segmentation on the vibration signal to obtain at least one signal segment.
Specifically, the step of performing hierarchical segmentation on the vibration signal includes:
after the vibration signal x (t) is obtained, determining a maximum allowable level number Nlevel according to the signal length N of the vibration signal, where the maximum allowable level number may be calculated in the following manner: log (log) 2 (N) -7. The maximum number of allowed tiers may be limited to 7.
And determining the actual layering number Nlevel according to the signal length N of the vibration signal and the maximum allowable layering number Nlevel. Calculation method 1: log of 2 (N) 0.75+1; calculation mode 2: n0.75. In general, the actual number of hierarchies Nlevel is the maximum value of the results in the calculation method 1 and the calculation method 2, and if the maximum value of the results in the calculation method 1 and the calculation method 2 is greater than the maximum allowable number of hierarchies Nlevel, the actual number of hierarchies Nlevel is the minimum value of the results in the calculation method 1 and the calculation method 2.
And layering the vibration signals according to the actual layering number nlevel to obtain at least one layer of signals.
And segmenting each layer of signal in the at least one layer of signal to obtain at least one signal segment. The specific number of segments is calculated as follows: 2. [ 3.
And S003, processing each signal segment in the at least one signal segment based on a signal separation operator SSO to obtain the kurtosis of each signal segment. Specifically, as shown in fig. 2, this step includes steps S0031 to S0033.
And S0031, performing signal separation operator SSO calculation on each signal segment in the at least one signal segment to obtain a frequency set and an amplitude set of each signal segment in each time.
Specifically, a Signal Separation Operator (SSO) is performed on each segment of the signal:
where t is time and θ is the set of frequencies and amplitude values at time t, which belong to the set
Refer to
The quotient space of (a); time t belonging to a set of real numbers
h is a window function, meaning h is a function of
Equal characteristics, a denotes the window width, k 1 =1, \8230; n is the signal length of the vibration signal;
represents a set of integers, represents k 1 Only integers can be taken;
represents cos (k) 1 θ)+isin(k 1 θ), i is a complex unit, δ is a sampling distance adjustable according to instantaneous frequency separation requirements;
the window width should be large enough, but for this solution, we need to calculate kurtosis for non-stationary processes, which means that the window width cannot be too long. Therefore, a balance needs to be struck between the accuracy of the SSO output of the signal separation operator and the accuracy of the spectral kurtosis calculation, and it is verified that the value of the parameter a should be 2 or less 7 。
Based on the signal separation operator, the time t of each signal segment and the corresponding frequency and amplitude can be output.
And S0032, determining the corresponding instantaneous frequency of each signal segment in each time according to the frequency set and the amplitude set of each signal segment in each time.
Specifically, the step is to perform thresholding on the frequency set at each time and then perform clustering to obtain at least one disjoint frequency cluster at each time; and then determining the frequency corresponding to the maximum amplitude value in each frequency cluster of at least one frequency cluster in each time, and taking the frequency as the instantaneous frequency of the moment.
An appropriate threshold value mu can be set for the output of the signal separation operator SSO
(t, theta) thresholding
Where μ is set to μ>0。
The thresholded frequencies are clustered by optimizing the choice of a parameter η determined by the time variation of t, and converted into 1 or more non-empty frequency clusters
k 2 =1, \8230, n, n is the number of frequency clusters. The parameter η is preferably set to 0.01.
The frequency corresponding to the maximum amplitude in each of the at least one frequency cluster per time,
taking the frequency as the instantaneous frequency of the moment
Step S0033, calculating the absolute value of the SSO output result to obtain the instantaneous amplitude of the instantaneous frequency
And S0034, calculating the complex envelope of the instantaneous frequency and the instantaneous amplitude of all the time of each signal segment to obtain the kurtosis of each signal segment. In particular, define
(t, theta) has a 2 p-order spectral distance of
p is a natural number greater than 0, and when p is 1 and 2 respectively, the kurtosis spectrum of each signal based on signal separation operator is obtained
And S004, determining a signal segment with the highest kurtosis in all the signal segments, and constructing a band-pass filter according to the center frequency and the frequency resolution of the signal segment with the highest kurtosis.
Specifically, a frequency matching structure diagram is established according to the kurtosis of each signal segment obtained in the above steps and the corresponding center frequency, and the obtained result is shown in fig. 4; or, establishing a frequency resolution matching structure chart according to the kurtosis of each signal segment and the corresponding frequency resolution;
and determining the signal segment with the highest kurtosis in all the signal segments according to the frequency matching structure chart or the frequency resolution matching structure chart, taking the frequency matching structure of fig. 4 as an example, namely the brightest part in a virtual coil in the diagram, and constructing a band-pass filter according to the central frequency and the frequency resolution of the signal segment.
And S005, filtering the vibration signal by using the band-pass filter.
And S006, obtaining fault characteristic frequency based on the filtered vibration signal, and identifying a fault state.
Specifically, complex envelope analysis is performed on the filtered signal, and the fault characteristic frequency is extracted, so as to obtain a result shown in fig. 5. And then, the bearing fault state is identified by combining the size and the running speed of the bearing, and the bearing retainer fault is diagnosed.
The second aspect of the invention discloses a train bearing fault feature precision extraction and diagnosis device, as shown in fig. 6, comprising a signal acquisition unit, a signal layering and segmentation unit, a kurtosis calculation unit, a band-pass filter construction unit and a fault state identification unit which are connected in sequence by signals,
the signal acquisition unit is used for acquiring a vibration signal of a bearing to be detected of the train, and specifically, the signal acquisition unit can be a wired signal receiving unit or a wireless signal receiving unit so as to receive an output signal of an acceleration sensor or other sensors.
The signal layering and segmenting unit is used for layering and segmenting the vibration signal to obtain at least one signal segment;
the kurtosis calculating unit is used for processing each signal segment in the at least one signal segment based on a signal separation operator SSO to obtain the kurtosis of each signal segment;
the band-pass filter constructing unit is used for determining a signal segment with the highest kurtosis in all the signal segments, constructing a band-pass filter according to the center frequency and the frequency resolution of the signal segment with the highest kurtosis, and filtering the vibration signal by using the band-pass filter;
the fault state identification unit is used for obtaining fault characteristic frequency based on the filtered vibration signals and identifying fault states.
The invention provides a train bearing fault feature precise extraction and diagnosis device, which comprises a memory and a controller which are sequentially connected in a communication manner, wherein the memory is stored with a computer program, and the processor is used for reading the computer program and executing the train bearing fault feature precise extraction and diagnosis method in the first aspect and any one possibility. For example, the Memory may include, but is not limited to, a Random-Access Memory (RAM), a Read-Only Memory (ROM), a Flash Memory (Flash Memory), a First-in First-out (FIFO), a First-in First-out (First Input Last Output, FILO), and/or a First-in Last-out (First Input Last Output, FILO), etc.; the processor may not be limited to the use of a microprocessor of the model number STM32F105 family. Furthermore, the computer device may also include, but is not limited to, a power supply unit, a display screen, and other necessary components.
A fourth aspect of the present invention provides a computer-readable storage medium, which stores instructions that, when executed on a computer, perform the method for precisely extracting and diagnosing train bearing fault features according to the first aspect and any one of the possibilities.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: modifications may be made to the embodiments described above, or equivalents may be substituted for some of the features described. And such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (6)
1. A train bearing fault feature precise extraction and diagnosis method is characterized by comprising the following steps:
acquiring a vibration signal of a bearing to be detected of a train;
carrying out layered segmentation on the vibration signal to obtain at least one signal segment;
processing each signal segment in the at least one signal segment based on a signal separation operator SSO to obtain the kurtosis of each signal segment;
determining a signal segment with the highest kurtosis in all signal segments, and constructing a band-pass filter according to the center frequency and the frequency resolution of the signal segment with the highest kurtosis;
filtering the vibration signal by using the band-pass filter;
obtaining fault characteristic frequency based on the filtered vibration signal, and identifying a fault state;
the determining a signal segment with the highest kurtosis in all signal segments comprises:
establishing a frequency matching structure chart according to the kurtosis of each signal segment and the corresponding center frequency, or establishing a frequency resolution matching structure chart according to the kurtosis of each signal segment and the corresponding frequency resolution;
determining a signal segment with the highest kurtosis in all signal segments according to the frequency matching structure chart or the frequency resolution matching structure chart;
the processing each signal segment in at least one signal segment based on the signal separation operator SSO to obtain the kurtosis of each signal segment includes:
performing signal separation operator SSO calculation on each signal segment in the at least one signal segment to obtain a frequency set and an amplitude set of each signal segment in each time;
determining the instantaneous frequency corresponding to each time of each signal segment according to the frequency set and the amplitude set of each time of each signal segment;
calculating the absolute value of the SSO output result to obtain the instantaneous amplitude of the instantaneous frequency;
calculating the complex envelope of the instantaneous frequency and the instantaneous amplitude of each signal segment at all times to obtain the kurtosis of each signal segment;
the determining the instantaneous frequency corresponding to each time of each signal segment according to the frequency set and the amplitude set of each time of each signal segment comprises:
clustering the frequency sets in each time after thresholding to obtain at least one disjoint frequency cluster in each time;
and determining the frequency corresponding to the maximum amplitude value in each frequency cluster of at least one frequency cluster in each time, and taking the frequency as the instantaneous frequency of the moment.
2. The method for precisely extracting and diagnosing the fault characteristics of the train bearing according to claim 1, wherein the window width of the signal separation operator SSO is less than 2 7 。
3. The train bearing fault feature precise extraction and diagnosis method as claimed in claim 1, wherein the step of performing layered segmentation on the vibration signal to obtain at least one signal segment comprises:
determining the maximum allowable layering number according to the signal length of the vibration signal;
determining the actual layering number according to the signal length and the maximum allowable layering number of the vibration signal;
layering the vibration signals according to the actual layering number to obtain at least one layer of signals;
and segmenting each layer of signal in the at least one layer of signal to obtain at least one signal segment.
4. A train bearing fault feature precise extraction and diagnosis device is characterized by comprising a signal acquisition unit, a signal layering and segmenting unit, a kurtosis calculation unit, a band-pass filter construction unit and a fault state identification unit which are sequentially connected by signals,
the signal acquisition unit is used for acquiring a vibration signal of a bearing to be detected of the train;
the signal layering and segmenting unit is used for layering and segmenting the vibration signal to obtain at least one signal segment;
the kurtosis calculating unit is used for processing each signal segment in the at least one signal segment based on a signal separation operator SSO to obtain the kurtosis of each signal segment;
the band-pass filter constructing unit is used for determining a signal segment with the highest kurtosis in all signal segments, constructing a band-pass filter according to the center frequency and the frequency resolution of the signal segment with the highest kurtosis, and filtering the vibration signal by using the band-pass filter;
the fault state identification unit is used for obtaining fault characteristic frequency based on the filtered vibration signal and identifying a fault state;
when the band-pass filter constructing unit is configured to determine a signal segment with a highest kurtosis among all signal segments, the band-pass filter constructing unit is specifically configured to:
establishing a frequency matching structure chart according to the kurtosis of each signal segment and the corresponding center frequency, or establishing a frequency resolution matching structure chart according to the kurtosis of each signal segment and the corresponding frequency resolution;
determining a signal segment with the highest kurtosis in all signal segments according to the frequency matching structure chart or the frequency resolution matching structure chart;
the kurtosis calculating unit is specifically configured to, when being configured to process each signal segment of the at least one signal segment based on the signal separation operator SSO to obtain a kurtosis of each signal segment:
performing signal separation operator SSO calculation on each signal segment in the at least one signal segment to obtain a frequency set and an amplitude set of each signal segment in each time;
determining the instantaneous frequency corresponding to each time of each signal segment according to the frequency set and the amplitude set of each time of each signal segment;
calculating the absolute value of the SSO output result to obtain the instantaneous amplitude of the instantaneous frequency;
calculating the complex envelope of the instantaneous frequency and the instantaneous amplitude of each signal segment at all times to obtain the kurtosis of each signal segment;
the kurtosis calculating unit, when configured to determine an instantaneous frequency corresponding to each time of each signal segment according to the frequency set and the amplitude set of each time of each signal segment, is specifically configured to:
clustering the frequency sets in each time after thresholding to obtain at least one disjoint frequency cluster in each time;
and determining the frequency corresponding to the maximum amplitude in each frequency cluster of at least one frequency cluster at each time, and taking the frequency as the instantaneous frequency at the moment.
5. The utility model provides a train bearing fault characteristic precision extraction and diagnostic equipment which characterized in that, includes memory and treater of communication connection in proper order, the computer program is stored on the memory, its characterized in that: the processor is used for reading the computer program and executing the train bearing fault characteristic precise extraction and diagnosis method of any one of claims 1 to 3.
6. A computer-readable storage medium having instructions stored thereon, characterized in that: when the instructions are run on a computer, the method for precisely extracting and diagnosing the train bearing fault characteristics according to any one of claims 1 to 3 is executed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110846485.6A CN113567128B (en) | 2021-07-26 | 2021-07-26 | Method, equipment and storage medium for precisely extracting and diagnosing fault characteristics of train bearing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110846485.6A CN113567128B (en) | 2021-07-26 | 2021-07-26 | Method, equipment and storage medium for precisely extracting and diagnosing fault characteristics of train bearing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113567128A CN113567128A (en) | 2021-10-29 |
| CN113567128B true CN113567128B (en) | 2023-03-14 |
Family
ID=78167585
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110846485.6A Active CN113567128B (en) | 2021-07-26 | 2021-07-26 | Method, equipment and storage medium for precisely extracting and diagnosing fault characteristics of train bearing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113567128B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102890143A (en) * | 2012-10-19 | 2013-01-23 | 西南交通大学 | Rail local defect vehicle-mounting detection method merging with vehicle speed information and front and rear axle box acceleration information |
| CN106248356A (en) * | 2016-07-13 | 2016-12-21 | 广东工业大学 | A kind of rotary machinery fault diagnosis method based on kurtosis index |
| CN109030001A (en) * | 2018-10-08 | 2018-12-18 | 电子科技大学 | A kind of Fault Diagnosis of Roller Bearings based on improvement HHT |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1579608A4 (en) * | 2002-10-11 | 2012-09-05 | Flint Hills Scient Llc | Method, computer program, and system for intrinsic timescale decomposition, filtering, and automated analysis of signals of arbitrary origin or timescale |
| US7706478B2 (en) * | 2005-05-19 | 2010-04-27 | Signalspace, Inc. | Method and apparatus of source separation |
| CN105424364A (en) * | 2015-11-09 | 2016-03-23 | 北京交通大学 | Diagnostic method and device of train bearing failure |
| EP3309530A1 (en) * | 2016-10-11 | 2018-04-18 | ABB Schweiz AG | Detection of bearing fault |
| US11506570B2 (en) * | 2018-08-27 | 2022-11-22 | Hyundai Motor Company | Method for sensing damage of bearing of engine using vibration signal |
| CN109359619A (en) * | 2018-10-31 | 2019-02-19 | 浙江工业大学之江学院 | A kind of high density surface EMG Signal Decomposition Based method based on convolution blind source separating |
| AU2020103681A4 (en) * | 2020-11-26 | 2021-02-04 | Anhui University Of Technology | Rolling Bearing Fault Diagnosis Method Based on Fourier Decomposition and Multi-scale Arrangement Entropy Partial Mean Value |
-
2021
- 2021-07-26 CN CN202110846485.6A patent/CN113567128B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102890143A (en) * | 2012-10-19 | 2013-01-23 | 西南交通大学 | Rail local defect vehicle-mounting detection method merging with vehicle speed information and front and rear axle box acceleration information |
| CN106248356A (en) * | 2016-07-13 | 2016-12-21 | 广东工业大学 | A kind of rotary machinery fault diagnosis method based on kurtosis index |
| CN109030001A (en) * | 2018-10-08 | 2018-12-18 | 电子科技大学 | A kind of Fault Diagnosis of Roller Bearings based on improvement HHT |
Non-Patent Citations (2)
| Title |
|---|
| 强背景噪声下列车轴承振动信号故障特征分离方法研究;刘文朋;《中国优秀博硕士学位论文全文数据库(硕士)工程科技Ⅱ辑》;20190315(第03期);全文 * |
| 滚动轴承故障诊断及趋势预测方法研究;汤杰;《中国优秀博硕士学位论文全文数据库(硕士)工程科技Ⅱ辑》;20190115(第01期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113567128A (en) | 2021-10-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108195587B (en) | Motor rolling bearing fault diagnosis method and system | |
| CN103852271B (en) | High-speed train running gear fault diagnosis and remote monitoring system based on Internet of Things | |
| CN106441893B (en) | Train rolling bearing fault vibrates differentiating method with impurity | |
| CN108760327B (en) | Diagnosis method for rotor fault of aircraft engine | |
| CN109782603A (en) | The detection method and monitoring system of rotating machinery coupling fault | |
| CN106441896A (en) | Characteristic vector extraction method for rolling bearing fault mode identification and state monitoring | |
| CN208125377U (en) | A kind of motor rolling bearing failure diagnosis system | |
| CN110008898B (en) | Industrial equipment data edge processing method based on symbol and convolutional neural network | |
| CN109583386A (en) | A kind of intelligent rotating mechanical breakdown depth network characterization discrimination method | |
| CN110940518B (en) | Aerospace transmission mechanism analysis method based on fault data | |
| CN106596149A (en) | Method for monitoring and diagnosing flat wheel state of rail vehicle based on VMD | |
| CN111071291B (en) | Train wheel set monitoring system and train wheel set monitoring method | |
| CN108267312B (en) | A kind of subway train bearing intelligent diagnostic method based on fast search algorithm | |
| CN111678699B (en) | Early fault monitoring and diagnosing method and system for rolling bearing | |
| CN107424626A (en) | A kind of gantry crane trolley track sound monitoring and fault finding system | |
| CN109406147A (en) | A kind of train bearing rail side acoustics diagnostic method under speed change working condition | |
| CN111504647A (en) | AR-MSET-based performance degradation evaluation method for rolling bearing | |
| CN110631832A (en) | Subway vehicle bearing fault on-line detection method | |
| CN114462480A (en) | Multi-source sensor rolling mill fault diagnosis method based on non-equilibrium data set | |
| CN113567128B (en) | Method, equipment and storage medium for precisely extracting and diagnosing fault characteristics of train bearing | |
| CN112113766B (en) | Characteristic extraction method for early damage state of rolling bearing | |
| CN106596103B (en) | Device for bearing detection on wheel-free lathe and bearing detection method | |
| CN116685967A (en) | Method and device for detecting anomalies in a machine or machine component | |
| CN116644278A (en) | Sample screening and feature extraction-based dual-model bearing anomaly detection algorithm | |
| CN110057587A (en) | A kind of nuclear power pump bearing intelligent failure diagnosis method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |