CN111060314A - Fault diagnosis method and test simulation device for rolling bearing of motor train unit - Google Patents
Fault diagnosis method and test simulation device for rolling bearing of motor train unit Download PDFInfo
- Publication number
- CN111060314A CN111060314A CN201911160769.9A CN201911160769A CN111060314A CN 111060314 A CN111060314 A CN 111060314A CN 201911160769 A CN201911160769 A CN 201911160769A CN 111060314 A CN111060314 A CN 111060314A
- Authority
- CN
- China
- Prior art keywords
- vibration
- bearing
- fault
- sequence
- axle box
- 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.)
- Granted
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
Abstract
The invention provides a fault diagnosis method and a test simulation device for a rolling bearing of a motor train unit. The improved multi-scale permutation entropy IMPE has low length dependence on signals and strong stability and robustness; enough useful information can be fully obtained in the coarse graining process, and compared with the traditional method, the method has strong superiority; the extracted characteristic values are input into the established LIBSVM model, the fault position of the bearing and the fault degree of the bearing can be effectively identified, and the diagnosis efficiency and the accuracy are high.
Description
Technical Field
The invention relates to the technical field of motor train unit fault diagnosis, in particular to a motor train unit rolling bearing fault diagnosis method and a test simulation device.
Background
The bearing fault diagnosis of the high-speed train is mainly based on vibration signal analysis. Firstly, a sensor technology is used for collecting vibration signals of a train axle box bearing, and then characteristic extraction is carried out on the bearing signals.
The traditional fault feature extraction method comprises a chaotic model, a time domain analysis method, a frequency domain analysis method and the like. However, a large amount of calculation time is required when extracting the signal characteristic value, and the vibration signal often has a nonlinear characteristic due to clearance change, load change, speed change and other reasons in the running process of the bearing, so that some inherent limitations exist when applying the traditional method.
With the development of nonlinear dynamics, methods based on entropy theory are applied to bearing fault diagnosis, and common methods include approximate entropy, sample entropy, permutation entropy, multi-scale permutation entropy and the like, and enough information with fault characteristics can be obtained from vibration signals by applying the methods. And finally, the health state of the tested bearing can be effectively identified by using a support vector machine technology. However, due to the fact that methods such as approximate entropy, sample entropy and permutation entropy have a single scale, some information is lost when a vibration signal characteristic value is extracted, and the identification accuracy is not high when fault diagnosis is performed. Although the multi-scale entropy and the multi-scale permutation entropy overcome the defect of single scale, the coarse graining process of the multi-scale entropy and the multi-scale permutation entropy on the vibration signal during the characteristic value extraction can shorten the time sequence length, so that the information is lost, and the fault identification accuracy is reduced.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides the fault diagnosis method and the test simulation device for the rolling bearing of the motor train unit, which have good simulation effect and high fault analysis accuracy.
The technical scheme of the invention is as follows: a fault diagnosis method for a rolling bearing of a motor train unit comprises the following steps:
s1), collecting a vibration acceleration signal sample of the fault bearing through a vibration acceleration sensor, and intercepting a signal with the length of N from the collected signal sample as a training sample, namely a vibration signal original time sequence;
s2), extracting characteristic values of the training samples by using improved multi-scale permutation entropy IMPE, wherein the improved multi-scale permutation entropy IMPE has the following calculation formula:
wherein x, tau, m and L are the original time sequence, scale factor, embedding dimension and time delay respectively,
wherein the content of the first and second substances,
wherein the content of the first and second substances,
is tau number of p(τ)(π) equalized value, p(τ)And (pi) is the frequency of appearance of arrangement pi, pi is the numbered arrangement of m-element vectors in the coarse grained time sequence, and the coarse grained time sequence is a sequence formed by performing coarse grained calculation on the original time sequence.
Extracting characteristic values by using improved multi-scale permutation entropy IMPE, which specifically comprises the following steps:
s201), carrying out coarse graining treatment on the vibration signal original time sequence x with the length of N:
when the scale factor tau is 1, the coarse grained time sequence is the original time sequence;
when the scale factor tau is more than or equal to 2, carrying out coarse graining treatment on the original time sequence, wherein k (k is more than or equal to 1 and less than or equal to tau) coarse graining signal sequences exist;
starting from the (j-1) th τ + k element of the original time series when the kth (1 ≦ k ≦ τ) coarse physicochemical series is calculated, wherein,
taking tau elements to calculate the mean value, and recording as
Namely, it is
1≤k≤τ；
The calculated mean constitutes a new time series, i.e. a coarse grained time series
S202), under the scale factor tau, the total number of the coarse graining time sequences is tau, for each coarse graining time sequence, every L-1 elements from j elements, m elements are taken as m-element vectors in sequence and are recorded as vectors
S203), for each vector
The elements in the vector are numbered from 0 to m-1 in sequence, and then each element is arranged from small to large in sequence to obtain a new arrangement, and each element in the new arrangement is arranged in the vector
Corresponding elements are corresponding to the elements, the corresponding numbers of each element are recorded according to the new arrangement element sequence, the corresponding numbers form a new sequence pi, and the total number m! Calculating the occurrence frequency of each type of sequence pi to obtain pn (τ)(pi), wherein n is more than or equal to 1 and less than or equal to m! ,
s204), at scale factor τ, the frequency p of occurrence of each type of sequence in step S203) because there are τ coarse-grained time sequencesn (τ)Having τ (pi), averaging to obtain
Obtaining a characteristic value under the scale factor tau by applying an IMPE calculation formula;
s3), repeating the steps to obtain tau characteristic values to form a characteristic set;
s4), inputting the feature set in the step S3) into a support vector machine as an input sample, and establishing a LIBSVM model;
s5), processing the sample to be detected according to S1-S2), and inputting the obtained feature set of the detected bearing into the trained LIBSVM model for classification to obtain the fault position and the fault degree of the bearing.
Preferably, the invention further provides a simulation device for fault diagnosis of the rolling bearing of the motor train unit, and the simulation device is used for simulating the fault bearing so as to obtain a training sample. The simulation device comprises a base and a liftable three-coordinate type electromagnetic vibration device arranged on the base, wherein the electromagnetic vibration device can generate vertical vibration and can also provide transverse and longitudinal vibration according to needs.
The electromagnetic vibration device on install the truss-like centre gripping frock that forms by the concatenation of a plurality of channel-section steels, the centre gripping frock on install radial loading device, the mounting bolt hole position that corresponding position set up, size and vertical rigidity on centre gripping frock and the radial loading device suit with the mounting bolt hole of experimental axle box, the structure of experimental axle box is unanimous with the structure of being surveyed high-speed train axle box, its inside bearing sets up to the fault bearing.
Preferably, the base is connected with the electromagnetic vibration device through an air spring and a vertical vibration absorber, the air leakage port and the air inlet of the air spring are respectively connected with an air leakage control valve and an air inlet control valve, and the air inlet control valve is connected with the electric air pump. Before simulation, the vertical rigidity of the air spring is the same as that of the measured high-speed train suspension by controlling the air relief control valve, the air inlet control valve and the electric air pump, and the damping coefficient of the vertical shock absorber is adjusted to be the same as that of the measured high-speed train suspension.
Preferably, the radial loading device is mounted on the truss through a bolt, the radial loading device is used for applying a specific pressure to the test axle box, meanwhile, the radial loading device is connected with the axle box, and the radial loading device is controlled to output a specific radial loading force, so that the loading pressure applied to the axle box is the same as the pressure applied to the axle box of the measured high-speed train.
Preferably, a bearing having a fault is mounted in the axle box, and a vibration acceleration sensor is attached to an outer ring of the bearing having a fault; the input shaft of the axle box is connected with the output end of a driving motor, a rotating speed sensor and a torque sensor are arranged on the output end of the driving motor, and the driving motor is rigidly connected with a motor support through a bolt.
Preferably, the faulty bearing is set as a bearing with different fault positions and different fault degrees, and includes a rolling body, an inner ring and an outer ring, and the fault bearing can be set to different damage degrees, which are respectively: early, medium and late stage.
The simulation device is used for simulating the vibration condition of the bearing when the high-speed train runs, the vibration acceleration signal of the bearing is obtained through the vibration acceleration sensor and is used as a training sample, the characteristic value is extracted through the improved multi-scale permutation entropy IMPE, the LIBSVM model is trained through the characteristic value, and the fault condition of the high-speed train bearing is identified and diagnosed through the trained model.
Preferably, the radial loading device comprises an integrated hydraulic pump station, a piston rod and a cross beam, the hydraulic pump station is installed on the truss through a bolt, the piston rod can move upwards or downwards in the hydraulic pump station, the cross beam is fixed on the piston rod, and the upper end of the cross beam is connected with the axle box. The central processor controls the hydraulic pump station to work, and the hydraulic pump station drives the piston rod to move upwards to apply specific loading pressure to the axle box through the cross beam.
Preferably, the electromagnetic vibration device is a triaxial vibration table, three vibration exciters are arranged in the triaxial vibration table, the three vibration exciters are respectively installed along the X-axis direction, the Y-axis direction and the Z-axis direction, the vibration exciters in the Z-axis direction mainly generate vertical vibration to simulate vertical vibration in the running process of the motor train unit, and the vibration exciters in the X-axis direction and the Y-axis direction respectively generate transverse vibration and longitudinal vibration to simulate transverse vibration and longitudinal vibration in the running process of the motor train unit.
The invention has the beneficial effects that:
1. the method utilizes the improved multi-scale permutation entropy IMPE to extract the characteristic value of the signal, has low length dependence on the signal and has strong stability and robustness;
2. enough useful information can be fully acquired in the coarse graining process, and the method has strong superiority compared with the traditional method, sample entropy, arrangement entropy, multi-scale entropy and the like;
3. the invention inputs the extracted characteristic value into the established LIBSVM model, can effectively identify the fault position of the bearing and the fault degree of the bearing, and has high diagnosis efficiency and high accuracy.
Drawings
FIG. 1 is a schematic structural diagram of a simulation apparatus according to embodiment 1 of the present invention;
FIG. 2 is a flowchart of a diagnostic method according to embodiment 2 of the present invention;
FIG. 3 is a waveform diagram of vibration acceleration signals of a test bearing according to embodiment 3 of the present invention;
FIG. 4 is a waveform diagram of vibration acceleration signals of a measured bearing according to embodiment 3 of the present invention;
FIG. 5 is a diagram of the coarse granulation process of the original time series in example 3 of the present invention
In the figure, 1-a base, 2-an electromagnetic vibration device, 3-a clamping tool, 4-a radial loading device, 5-a test axle box, 7-a central processing unit, 11-an air spring, 12-a vertical vibration absorber, 13-a gas release control valve, 14-an air inlet control valve, 15-an electric air pump, 51-a vibration acceleration sensor, 61-a motor support, 62-a driving motor, 63-a rotating speed sensor and 64-a torque sensor;
Detailed Description
The following further describes embodiments of the present invention with reference to the accompanying drawings:
example 1
As shown in fig. 1, the embodiment provides a simulation device for diagnosing a fault of a rolling bearing of a motor train unit, and the simulation device simulates the fault bearing to obtain a training sample. The simulation device comprises a base 1 and a liftable three-coordinate type electromagnetic vibration device 2 arranged on the base 1, wherein the electromagnetic vibration device 2 can generate vertical vibration and can also provide transverse and longitudinal vibration according to needs.
A bearing with a fault is installed in the test axle box 5, and a vibration acceleration sensor 51 is adhered to the outer ring of the bearing with the fault; the input shaft of the test axle box 5 is connected with the output end of the driving motor 62.
Preferably, the output end of the driving motor 62 is provided with a rotation speed sensor 63 and a torque sensor 64, and the driving motor 62 is rigidly connected with the motor support 61 through a bolt.
Preferably, the base 1 is connected with the electromagnetic vibration device 2 through an air spring 11 and a vertical shock absorber 12, an air leakage port and an air inlet of the air spring 11 are respectively connected with an air leakage control valve 13 and an air inlet control valve 14, and the air inlet control valve 13 is connected with an electric air pump 15. Before simulation, the central processing unit 7 controls the air discharge control valve 13, the air inlet control valve 14 and the electric air pump 15, so that the vertical rigidity of the air spring 11 is the same as that of the measured high-speed train suspension, and simultaneously, the damping coefficient of the vertical shock absorber 12 is adjusted, so that the damping coefficient is the same as that of the measured high-speed train suspension.
Preferably, the radial loading device comprises an integrated hydraulic pump station, a piston rod and a cross beam, the hydraulic pump station is installed on the truss through a bolt, the piston rod can move upwards or downwards in the hydraulic pump station, the cross beam is fixed on the piston rod, and the upper end of the cross beam is connected with the axle box. The central processor controls the hydraulic pump station to work, and the hydraulic pump station drives the piston rod to move upwards to apply specific loading pressure to the axle box through the cross beam.
Preferably, the electromagnetic vibration device is a triaxial vibration table, three vibration exciters are arranged in the triaxial vibration table, the three vibration exciters are respectively installed along the X-axis direction, the Y-axis direction and the Z-axis direction, the vibration exciters in the Z-axis direction mainly generate vertical vibration to simulate vertical vibration in the running process of the motor train unit, and the vibration exciters in the X-axis direction and the Y-axis direction respectively generate transverse vibration and longitudinal vibration to simulate transverse vibration and longitudinal vibration in the running process of the motor train unit.
Preferably, during sample simulation, the faulty bearing is set to be a bearing with different fault positions and different fault degrees, the fault position includes a rolling element, an inner ring and an outer ring, and the fault degree of the bearing can be set to be different damage degrees, which are respectively: early, medium and late stage.
During test simulation, a bearing in the test axle box 5 is set as a fault bearing, a vibration acceleration sensor 51 is pasted on the outer ring of the bearing, the central processing unit 7 controls the electromagnetic vibration device 2 to generate specific amplitude and frequency, and the electromagnetic vibration device 2 can generate vertical vibration and can also provide transverse and longitudinal vibration according to requirements; meanwhile, the central processing unit 7 controls the radial loading device 4 to apply a specific radial loading force to the test axle box 5; meanwhile, the central processing unit 7 also controls the driving motor 62 to output a specific rotating speed, the rotating speed sensor 63 detects the output rotating speed of the driving motor 62 to ensure that the output rotating speed is correct, and the torque sensor 64 detects the output torque change. Then, collecting a vibration acceleration signal sample of the fault bearing by using a vibration acceleration sensor 51; and (3) intercepting the signal with the length N from the collected signal sample to be used as a training sample, namely the original time sequence of the vibration signal.
Example 2
The embodiment provides a fault diagnosis method for a rolling bearing of a motor train unit, which comprises the following steps:
s1), collecting a vibration acceleration signal sample of the fault bearing through the vibration acceleration sensor 51, and intercepting a signal with the length of N from the collected signal sample as a training sample, namely a vibration signal original time sequence;
s2), extracting characteristic values of the training samples by using improved multi-scale permutation entropy IMPE, wherein the improved multi-scale permutation entropy IMPE has the following calculation formula:
wherein x, tau, m and L are the original time sequence, scale factor, embedding dimension and time delay respectively,
wherein the content of the first and second substances,
wherein the content of the first and second substances,
is tau number of p(τ)(π) equalized value, p(τ)And (pi) is the frequency of appearance of arrangement pi, pi is the numbered arrangement of m-element vectors in the coarse grained time sequence, and the coarse grained time sequence is a sequence formed by performing coarse grained calculation on the original time sequence.
Extracting characteristic values by using improved multi-scale permutation entropy IMPE, which specifically comprises the following steps:
s201), carrying out coarse graining treatment on the vibration signal original time sequence x with the length of N:
when the scale factor tau is 1, the coarse grained time sequence is the original time sequence;
when the scale factor tau is more than or equal to 2, carrying out coarse graining treatment on the original time sequence, wherein k (k is more than or equal to 1 and less than or equal to tau) coarse graining signal sequences exist;
starting from the (j-1) th τ + k element of the original time series when the kth (1 ≦ k ≦ τ) coarse physicochemical series is calculated, wherein,
taking tau elements to calculate the mean value, and recording as
Namely, it is
1≤k≤τ；
The calculated mean constitutes a new time series, i.e. a coarse grained time series
S202), under the scale factor tau, the total number of the coarse graining time sequences is tau, for each coarse graining time sequence, every L-1 elements from j elements, m elements are taken as m-element vectors in sequence and are recorded as vectors
S203), for each vector
The elements in the vector are numbered from 0 to m-1 in sequence, and then each element is arranged from small to large in sequence to obtain a new arrangement, and each element in the new arrangement is arranged in the vector
Corresponding elements are corresponding to the elements, the corresponding numbers of each element are recorded according to the new arrangement element sequence, the corresponding numbers form a new sequence pi, and the total number m! Calculating the occurrence frequency of each type of sequence pi to obtain pn (τ)(pi), wherein n is more than or equal to 1 and less than or equal to m! ,
s204), at scale factor τ, the frequency p of occurrence of each type of sequence in step S203) because there are τ coarse-grained time sequencesn (τ)(π) hasTau are obtained by averaging
Obtaining a characteristic value under the scale factor tau by applying an IMPE calculation formula;
s3), repeating the steps to obtain tau characteristic values to form a characteristic set;
s4), inputting the feature set in the step S3) into a support vector machine as an input sample, and establishing a LIBSVM model;
s5), processing the sample to be detected according to S1-S3), and inputting the obtained feature set of the detected bearing into the trained LIBSVM model for classification to obtain the fault position and the fault degree of the bearing.
Example 3
Based on embodiments 1 and 2, this embodiment replaces the bearings in the test axle box 5 with failed bearings having rolling element failure (early and late), inner ring failure (early and late), outer ring failure (early and late) and normal state bearings for a plurality of times, respectively labeled as BE1, BE2, IR1, IR2, OR1, OR2, Norm 7 states;
after each replacement, the test axle box 5 is installed on the test bed, and step S1) is performed to intercept 25 groups of data as an original time sequence by using the signal with the length N4096 acquired by the vibration acceleration signal 51; when an improved multi-scale arrangement entropy IMPE method is adopted, x is an original signal sequence with the length of 4096, a scale factor tau is 20, an embedding dimension m is 5, and a time delay L is 1; fig. 3 shows waveforms of the acceleration vibration signals collected in the 7 states.
Performing steps S2-S3) on 25 groups of original signal sequences, inputting the obtained 25 groups of characteristic values into a support vector machine, and establishing an LIBSVM model;
then placing a tested bearing with unknown fault in the tested shaft box, and executing the steps S2-S3) to obtain an acceleration vibration signal waveform diagram as shown in FIG. 4; the coarse granulation process of step b is shown in fig. 5.
And inputting the feature set obtained by the calculation in the step S3) into the trained LIBSVM model to classify the state of the LIBSVM model, wherein the classification result is shown in table 1.
TABLE 1 results of the classification
According to the classification results shown in table 1, the obtained vibration acceleration signals of 25 groups of tested bearings all indicate that the failure mode of the tested bearing in the tested axle box is outer ring early failure damage, according to the process, the bearings with rolling body failure, inner ring failure and outer ring failure are placed in the experimental axle box 5, and the steps from S1 to S3 are repeated, so that the running state of the bearing in the tested axle box can be accurately detected by using the trained LIBSVM model.
The foregoing embodiments and description have been presented only to illustrate the principles and preferred embodiments of the invention, and various changes and modifications may be made therein without departing from the spirit and scope of the invention as hereinafter claimed.
Claims (10)
1. A fault diagnosis method for a rolling bearing of a motor train unit is characterized by comprising the following steps:
s1), collecting a vibration acceleration signal sample of the fault bearing through a vibration acceleration sensor, and intercepting a signal with the length of N from the collected signal sample as a training sample, namely a vibration signal original time sequence;
s2), extracting characteristic values of the training samples by using improved multi-scale permutation entropy IMPE, wherein the improved multi-scale permutation entropy IMPE has the following calculation formula:
wherein x, tau, m and L are the original time sequence, scale factor, embedding dimension and time delay respectively,
wherein the content of the first and second substances,
wherein the content of the first and second substances,
is tau number of p(τ)(π) equalized value, p(τ)(pi) is the frequency of occurrence of arrangement pi, pi is the numbering arrangement of m-element vectors in the coarse graining time sequence, and the coarse graining time sequence is a sequence formed by the coarse graining calculation of the original time sequence;
s3), repeating the step S2) to obtain tau characteristic values to form a characteristic set;
s4), inputting the feature set in the step S3) into a support vector machine as an input sample, and establishing a LIBSVM model;
s5), processing the sample to be detected according to S1-S2), and inputting the obtained feature set of the detected bearing into the trained LIBSVM model for classification to obtain the fault position and the fault degree of the bearing.
2. The method for diagnosing the fault of the rolling bearing of the motor train unit according to claim 1, wherein the method comprises the following steps: step S2), extracting feature values by using the improved multi-scale permutation entropy IMPE, specifically:
carrying out coarse graining treatment on the original time sequence x of the vibration signal with the length N:
when the scale factor τ is 1, the coarse grained time series is the original time series.
3. The method for diagnosing the fault of the rolling bearing of the motor train unit according to claim 2, wherein the method comprises the following steps: when the scale factor tau is more than or equal to 2, the method further comprises the following steps:
carrying out coarse graining treatment on the original time sequence, wherein k (k is more than or equal to 1 and less than or equal to tau) coarse graining signal sequences exist;
when the k (1 ≦ k ≦ τ) coarse physicochemical sequence is calculated, starting from the (j-1) τ + k element of the original time sequenceWherein, in the step (A),
taking tau elements to calculate the mean value, and recording as
Namely, it is
1≤k≤τ；
The calculated mean constitutes a new time series, i.e. a coarse grained time series
4. The method for diagnosing the fault of the rolling bearing of the motor train unit according to claim 3, wherein the method comprises the following steps: after obtaining the coarse grained time series
The method also comprises the following steps:
under the scale factor tau, tau coarse graining time sequences are totally arranged, for each coarse graining time sequence, from j elements, every L-1 elements, m elements are taken as m-element vectors in sequence and are recorded as vectors
For each vector
The elements in (1) are numbered from 0 to m-1 in sequence, and then each element is arranged from small to large in sequence to obtainTo a new arrangement in which each element is in a vector
Corresponding elements are corresponding to the elements, the corresponding numbers of each element are recorded according to the new arrangement element sequence, the corresponding numbers form a new sequence pi, and the total number m! Calculating the occurrence frequency of each type of sequence pi to obtain pn (τ)(pi), wherein n is more than or equal to 1 and less than or equal to m! ,
at a scale factor τ, because there are τ coarse-grained time series, the frequency p at which each type of series occursn (τ)Having τ (pi), averaging to obtain
And obtaining the characteristic value under the scale factor tau by using an IMPE calculation formula.
5. The utility model provides a EMUs antifriction bearing fault diagnosis's experimental analogue means which characterized in that: simulating the fault bearing through the simulation device to obtain a training sample;
the simulation device comprises a base and a liftable three-coordinate type electromagnetic vibration device arranged on the base, wherein the electromagnetic vibration device can generate vertical vibration and can also provide transverse and longitudinal vibration according to needs;
a truss type clamping tool formed by splicing a plurality of channel steels is arranged on the electromagnetic vibration device;
the clamping tool is provided with a radial loading device which is used for generating specific pressure,
the radial loading device is connected with the axle box, and the loading pressure borne by the test axle box is the same as the pressure borne by the axle box of the measured high-speed train by controlling the radial loading device to output a specific radial loading force;
a bearing with a fault is installed in the axle box, and a vibration acceleration sensor is adhered to the outer ring of the bearing with the fault;
the simulation device is used for simulating the vibration condition of the bearing when the high-speed train runs, the vibration acceleration signal of the bearing is obtained through the vibration acceleration sensor and is used as a training sample, the characteristic value is extracted through the improved multi-scale permutation entropy IMPE, the LIBSVM model is trained through the characteristic value, and the fault condition of the high-speed train bearing is identified and diagnosed through the trained model.
6. The test simulation device for fault diagnosis of the rolling bearing of the motor train unit according to claim 5, characterized in that: the position, the size and the vertical rigidity of a mounting bolt hole arranged at a corresponding position on the clamping tool and the radial loading device are matched with those of a mounting bolt hole of a test axle box, the structure of the test axle box is consistent with that of a tested high-speed train axle box, and a bearing in the test axle box is set as a fault bearing.
7. The test simulation device for fault diagnosis of the rolling bearing of the motor train unit according to claim 5, characterized in that: the base is connected with the electromagnetic vibration device through an air spring and a vertical vibration absorber;
the air leakage port and the air inlet of the air spring are respectively connected with an air leakage control valve and an air inlet control valve, and the air inlet control valve is connected with an electric air pump;
before simulation, the vertical rigidity of the air spring is the same as that of the measured high-speed train suspension by controlling the air relief control valve, the air inlet control valve and the electric air pump, and the damping coefficient of the vertical shock absorber is adjusted to be the same as that of the measured high-speed train suspension.
8. The test simulation device for fault diagnosis of the rolling bearing of the motor train unit according to claim 5, characterized in that:
the radial loading device comprises an integrated hydraulic pump station, a piston rod and a cross beam, the hydraulic pump station is installed on the truss through a bolt, the piston rod can move upwards or downwards in the hydraulic pump station, the cross beam is fixed on the piston rod, and the upper end of the cross beam is connected with the axle box;
the electromagnetic vibration device is a triaxial vibration table, three vibration exciters are arranged in the triaxial vibration table and are respectively installed along the X-axis direction, the Y-axis direction and the Z-axis direction, the vibration exciters in the Z-axis direction mainly generate vertical vibration to simulate the vertical vibration in the running process of the motor train unit, and the vibration exciters in the X-axis direction and the Y-axis direction respectively generate transverse vibration and longitudinal vibration to simulate the transverse vibration and the longitudinal vibration in the running process of the motor train unit.
9. The test simulation device for fault diagnosis of the rolling bearing of the motor train unit according to claim 5, characterized in that: the input shaft of the axle box is connected with the output end of a driving motor, a rotating speed sensor and a torque sensor are arranged on the output end of the driving motor, and the driving motor is rigidly connected with a motor support through a bolt.
10. The test simulation device for fault diagnosis of the rolling bearing of the motor train unit according to claim 5, characterized in that: trouble bearing set up to the bearing that has the position of different trouble and the fault degree of difference, the position of trouble includes rolling element, inner circle, outer lane, the fault degree of bearing can set up to different damage degrees, is respectively: early, medium and late stage.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911160769.9A CN111060314B (en) | 2019-11-23 | 2019-11-23 | Fault diagnosis method and test simulation device for rolling bearing of motor train unit |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911160769.9A CN111060314B (en) | 2019-11-23 | 2019-11-23 | Fault diagnosis method and test simulation device for rolling bearing of motor train unit |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111060314A true CN111060314A (en) | 2020-04-24 |
| CN111060314B CN111060314B (en) | 2021-10-26 |
Family
ID=70298591
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911160769.9A Active CN111060314B (en) | 2019-11-23 | 2019-11-23 | Fault diagnosis method and test simulation device for rolling bearing of motor train unit |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111060314B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113074942A (en) * | 2021-03-31 | 2021-07-06 | 国网北京市电力公司 | Monitoring device, bearing seat, air pump and transformer |
| WO2023137600A1 (en) * | 2022-01-18 | 2023-07-27 | 广州工商学院 | Variable-speed fault diagnosis apparatus for rolling bearing of high-speed locomotive |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009135639A1 (en) * | 2008-05-05 | 2009-11-12 | Aktiebolaget Skf | Determining hydrostatic leakage |
| CN104849050A (en) * | 2015-06-02 | 2015-08-19 | 安徽工业大学 | Rolling bearing fault diagnosis method based on composite multi-scale permutation entropies |
| CN108416268A (en) * | 2018-02-02 | 2018-08-17 | 华侨大学 | A kind of action identification method based on dual robot Visual Communication |
| CN109916628A (en) * | 2019-04-04 | 2019-06-21 | 哈尔滨理工大学 | Based on the Fault Diagnosis of Roller Bearings for improving multiple dimensioned amplitude perception arrangement entropy |
| CN110310660A (en) * | 2019-06-06 | 2019-10-08 | 上海工程技术大学 | A kind of voice re-sampling detection method based on sound spectrograph |
-
2019
- 2019-11-23 CN CN201911160769.9A patent/CN111060314B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009135639A1 (en) * | 2008-05-05 | 2009-11-12 | Aktiebolaget Skf | Determining hydrostatic leakage |
| CN104849050A (en) * | 2015-06-02 | 2015-08-19 | 安徽工业大学 | Rolling bearing fault diagnosis method based on composite multi-scale permutation entropies |
| CN108416268A (en) * | 2018-02-02 | 2018-08-17 | 华侨大学 | A kind of action identification method based on dual robot Visual Communication |
| CN109916628A (en) * | 2019-04-04 | 2019-06-21 | 哈尔滨理工大学 | Based on the Fault Diagnosis of Roller Bearings for improving multiple dimensioned amplitude perception arrangement entropy |
| CN110310660A (en) * | 2019-06-06 | 2019-10-08 | 上海工程技术大学 | A kind of voice re-sampling detection method based on sound spectrograph |
Non-Patent Citations (1)
| Title |
|---|
| 李永健: "高速列车轴箱轴承智能故障诊断技术研究", 《中国博士学位论文全文数据库 工程科技II辑》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113074942A (en) * | 2021-03-31 | 2021-07-06 | 国网北京市电力公司 | Monitoring device, bearing seat, air pump and transformer |
| WO2023137600A1 (en) * | 2022-01-18 | 2023-07-27 | 广州工商学院 | Variable-speed fault diagnosis apparatus for rolling bearing of high-speed locomotive |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111060314B (en) | 2021-10-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108426713B (en) | Rolling bearing weak fault diagnosis method based on wavelet transformation and deep learning | |
| CN111060314B (en) | Fault diagnosis method and test simulation device for rolling bearing of motor train unit | |
| CN110428004B (en) | Mechanical part fault diagnosis method based on deep learning under data imbalance | |
| CN107271187B (en) | The method that a kind of pair of automobile speed variator bearing failure carries out quantitative Diagnosis | |
| CN109781411A (en) | A kind of combination improves the Method for Bearing Fault Diagnosis of sparse filter and KELM | |
| CN110297479B (en) | Hydroelectric generating set fault diagnosis method based on convolutional neural network information fusion | |
| CN105606363B (en) | A kind of Method for Bearing Fault Diagnosis adaptive based on domain | |
| CN105738109A (en) | Bearing fault classification diagnosis method based on sparse representation and ensemble learning | |
| CN108956614A (en) | A kind of pit rope dynamic method for detection fault detection and device based on machine vision | |
| CN109211546A (en) | Rotary machinery fault diagnosis method based on noise reduction autocoder and incremental learning | |
| CN107199506B (en) | A kind of grinding trembling detection method based on stack self-encoding encoder and support vector machines | |
| CN113268833A (en) | Migration fault diagnosis method based on deep joint distribution alignment | |
| CN110530639A (en) | A kind of bullet train axle box bearing failure diagnostic method | |
| CN110084316A (en) | A kind of method for diagnosing faults based on the multiple dimensioned arrangement entropy and glowworm swarm algorithm Support Vector Machines Optimized of fine time shift | |
| CN116793682A (en) | Bearing fault diagnosis method based on iCORAL-MMD and anti-migration learning | |
| CN111595541A (en) | Multi-dimensional structure damage identification method based on mass vibration transmissibility data convolutional neural network processing | |
| Tao et al. | Fault diagnosis of rolling bearing using deep belief networks | |
| CN112101082A (en) | Rotary machine fault diagnosis method based on improved low-rank sparse decomposition | |
| CN111076934A (en) | Method for diagnosing potential fault of bearing based on S transformation | |
| CN109117896A (en) | A kind of Rolling Bearing Fault Character extracting method based on KSVD dictionary learning | |
| CN110057918B (en) | Method and system for quantitatively identifying damage of composite material under strong noise background | |
| CN107966099A (en) | A kind of detection method of part defective products | |
| CN114511521A (en) | Tire defect detection method based on multiple representations and multiple sub-field self-adaption | |
| CN106844991B (en) | Air spring rigidity self-balancing iterative identification method for air floating type vibration control system | |
| CN112146882B (en) | Bearing fault diagnosis method based on transfer learning vibration signal image recognition |
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 |