CN109708890B - Method for detecting the condition of bearings in a motor vehicle transmission - Google Patents
Method for detecting the condition of bearings in a motor vehicle transmission Download PDFInfo
- Publication number
- CN109708890B CN109708890B CN201811588388.6A CN201811588388A CN109708890B CN 109708890 B CN109708890 B CN 109708890B CN 201811588388 A CN201811588388 A CN 201811588388A CN 109708890 B CN109708890 B CN 109708890B
- Authority
- CN
- China
- Prior art keywords
- fault
- transmission
- alpha
- bearing
- bearings
- 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
- 230000005540 biological transmission Effects 0.000 title claims abstract description 84
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000012360 testing method Methods 0.000 claims abstract description 69
- 230000001133 acceleration Effects 0.000 claims abstract description 39
- 230000004927 fusion Effects 0.000 claims abstract description 19
- 230000009467 reduction Effects 0.000 claims abstract description 15
- 238000004458 analytical method Methods 0.000 claims abstract description 11
- 238000012549 training Methods 0.000 claims abstract description 7
- 238000000513 principal component analysis Methods 0.000 claims abstract description 6
- 238000012545 processing Methods 0.000 claims description 16
- FGUUSXIOTUKUDN-IBGZPJMESA-N C1(=CC=CC=C1)N1C2=C(NC([C@H](C1)NC=1OC(=NN=1)C1=CC=CC=C1)=O)C=CC=C2 Chemical compound C1(=CC=CC=C1)N1C2=C(NC([C@H](C1)NC=1OC(=NN=1)C1=CC=CC=C1)=O)C=CC=C2 FGUUSXIOTUKUDN-IBGZPJMESA-N 0.000 claims description 9
- 238000001228 spectrum Methods 0.000 claims description 9
- 239000006096 absorbing agent Substances 0.000 claims description 8
- 238000013016 damping Methods 0.000 claims description 8
- 229910000831 Steel Inorganic materials 0.000 claims description 6
- 238000000354 decomposition reaction Methods 0.000 claims description 6
- 239000010959 steel Substances 0.000 claims description 6
- 239000000725 suspension Substances 0.000 claims description 6
- 238000004422 calculation algorithm Methods 0.000 claims description 4
- 239000003638 chemical reducing agent Substances 0.000 claims description 4
- 230000001186 cumulative effect Effects 0.000 claims description 4
- 238000009434 installation Methods 0.000 claims description 4
- 238000012843 least square support vector machine Methods 0.000 claims description 4
- 238000005457 optimization Methods 0.000 claims description 4
- 239000002245 particle Substances 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000004891 communication Methods 0.000 claims description 3
- 239000006185 dispersion Substances 0.000 claims description 3
- 239000000203 mixture Substances 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims 1
- 238000010606 normalization Methods 0.000 claims 1
- 238000003745 diagnosis Methods 0.000 abstract description 14
- 230000000694 effects Effects 0.000 abstract description 4
- 238000000605 extraction Methods 0.000 abstract description 3
- 238000004088 simulation Methods 0.000 abstract description 2
- 230000008859 change Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000035939 shock Effects 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000002405 diagnostic procedure Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
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
Landscapes
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- General Physics & Mathematics (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention particularly relates to a method for detecting the state of each bearing in an automobile transmission, which has good simulation effect and accurate fault analysis. The vibration acceleration signal of the bearing is decomposed and reconstructed by an EMD method, so that the noise in the bearing signal of the automobile transmission can be effectively inhibited, and useful components in the signal are enhanced and highlighted; the signals subjected to EMD noise reduction are subjected to Alpha stable distribution and multi-fractal feature extraction respectively, and feature fusion is performed by utilizing nuclear principal component analysis, so that the respective advantages of the Alpha stable distribution and the multi-fractal are fully combined, and the precision and the efficiency of fault diagnosis are improved; the method has the advantages that relevant data of the tested gearboxes of the bearings with different fault degrees in the test bed are used as training samples to establish the PSO-LSSVM model, the relevant data of the tested gearboxes can be brought into the trained PSO-LSSVM model, accordingly, the fault positions and fault states of the bearings in the tested gearboxes are obtained through analysis, and the diagnosis efficiency and the accuracy are high.
Description
The application has the following application numbers: 201710678357.9, filing date: 2017-08-09, the patent name "a method for quantitatively diagnosing the bearing fault of the automobile transmission", and the patent name "are filed by divisional application.
Technical Field
The invention relates to a fault diagnosis technology of an automobile transmission, in particular to a method for detecting the state of each bearing in the automobile transmission.
Background
The rolling bearings used in the automobile transmission are various in types and large in number, for example, deep groove ball bearings, tapered roller bearings, double cylindrical roller bearings and the like, and various bearings in the transmission often run continuously in severe working conditions of speed change, heavy load and high temperature, so that fatigue damage is easy to occur. If the damage is not processed in time, the function of the bearing is completely failed, and a series of chain reactions are further caused, so that the whole automobile cannot work normally, serious economic loss is caused at low cost, and catastrophic casualties can be caused at high cost. At present, because the accidents caused by the fact that damage faults of automobile transmission bearings, such as pitting, cracks or scratches, are not found in time are frequent, monitoring and diagnosing the faults of the automobile transmission bearings are necessary.
In the prior art, the fault diagnosis research of the automobile transmission bearing is mainly based on vibration analysis, and common fault feature extraction methods such as statistical parameters, wavelet transformation, Wingelwiri distribution and the like have respective defects, so that the fault diagnosis method can cause the phenomena of misdiagnosis and missed diagnosis of faults due to unstable diagnosis results if applied to actual engineering; secondly, most of the existing research methods do not consider a plurality of factors in the actual running process of the automobile, such as influence of unsmooth lines, load change, speed change and the like on vibration signals of the automobile transmission bearing, so that the online monitoring effect of the automobile transmission bearing is not ideal. Therefore, it is necessary to design a test bed capable of accurately simulating the actual driving conditions of the automobile and analyze the test data by using a suitable analysis and diagnosis method to improve the precision of bearing fault diagnosis.
Disclosure of Invention
In view of the above problems, the present invention aims to provide a method for detecting the state of each bearing in an automotive transmission, which has good simulation effect and accurate fault analysis.
In order to achieve the purpose of the invention, the technical scheme adopted by the invention is as follows: the method for detecting the state of each bearing in the automobile transmission comprises a test transmission and a tested transmission, wherein the test transmission and the tested transmission can be installed on a bearing test bed, the bearing test bed comprises a base, a liftable three-coordinate type electromagnetic vibration device is arranged on the base, a truss type clamping tool formed by splicing a plurality of channel steel or rectangular steel is arranged on the electromagnetic vibration device, the position, the size and the vertical rigidity of an installation bolt hole arranged at a corresponding position on the clamping tool are matched with the installation bolt hole of the test transmission, the structure of the test transmission is consistent with that of the tested automobile transmission, and one or more fault bearings are arranged at a plurality of positions where the bearings are arranged in the test transmission;
the base is connected with the electromagnetic vibration device through an air spring and a vertical vibration absorber, an air inlet and an air outlet of the air spring are respectively connected with an air inlet control valve and an air release valve, the air inlet control valve is connected with an electric air pump, the vertical rigidity of the air spring is consistent with the vertical rigidity of the tested automobile suspension, and the damping coefficient of the vertical vibration absorber is consistent with the vertical damping coefficient of the tested automobile suspension;
the input shaft of the test transmission is connected with the output end of the torque loading device, the torque loading device comprises a motor, the output end of the motor is sequentially connected with the torque sensor and the input end of the fixed-gear-ratio reducer, and the output end of the fixed-gear-ratio reducer is connected with the input shaft of the test transmission; the output end of the motor is also provided with a rotating speed sensor;
an output shaft of the test transmission is connected with an inertia load device, the inertia load device comprises an auxiliary support which is mutually independent from the base, a transmission gear set is arranged on the auxiliary support, the transmission gear set is composed of a pair of cylindrical straight gears or a pair of conical gears, a driving gear of the transmission gear set is arranged on the output shaft of the test transmission, a driven gear of the transmission gear set is connected with a rotating wheel through an intermediate shaft, the rotating wheel is arranged on the auxiliary support through a bearing, and the auxiliary support is also provided with a hydraulic brake caliper which is adaptive to the size of the rotating wheel;
vibration acceleration sensors are pasted on outer rings of a plurality of bearings installed in the test transmission; the central processing unit is respectively in communication connection with a hydraulic cylinder control valve, a rotating speed sensor, a torque sensor, a motor, a gear shifting controller of a test transmission, a plurality of vibration acceleration sensors, an air inlet control valve, an air release valve, an electric air pump and an electromagnetic vibration device of the hydraulic brake caliper;
the method is characterized in that: the diagnostic method comprises the following steps:
setting bearings at one or more positions in a test transmission as fault bearings, pasting a vibration acceleration sensor on the outer ring of the fault bearing, and then sequentially carrying out the following steps:
a. the central processing unit controls the electromagnetic vibration device to generate specific amplitude and vibration frequency; meanwhile, the central processing unit controls the motor and the gear shifting controller to enable an output shaft of the test transmission to output a specific rotating speed; when the output shaft of the test speed changer outputs a specific rotating speed, the central processing unit controls the brake calipers to apply brake torque to the rotating wheel, so that the output shaft of the test speed changer is subjected to specific load torque; the method comprises the steps that a vibration acceleration sensor collects a vibration acceleration signal sample of a fault bearing; taking the collected sample as a training sample, and performing EMD self-adaptive decomposition on a vibration acceleration signal x (t) in the training sample, wherein the decomposition method comprises the following steps:
in the above formula, n is the number of decomposed IMF components; cjRepresents the jth IMF component, j 1, 2, 3.., n; r isnIs the residual component;
b. decomposing the mixture to obtain n CjAfter the components, each C is calculated separatelyi(j ═ 1, 2, 3.., n) kurtosis value, two of C with the highest kurtosis value and the second highest kurtosis value are selectediPerforming linear superposition to obtain an acceleration signal of the characteristic highlight subjected to EMD noise reduction, then averagely dividing the obtained acceleration signal of the characteristic highlight into m sections according to the time length, and recording the signals of different time length sections as S1-Sm;
c. respectively estimating Alpha stable distribution parameters of the S1-Sm sections in the step b and calculating probability density functions of the sections, and extracting 5 Alpha stable distribution characteristics of characteristic index Alpha (Alpha is more than 0 and less than or equal to 2), symmetric parameter beta (-beta is more than or equal to 1), dispersion coefficient gamma (gamma is more than 0), position parameter (-infinity is more than or equal to infinity) and extremum h (h is more than 0) of the probability density functions;
d. and (c) respectively carrying out multi-fractal detrending fluctuation analysis on the S1-Sm sections in the step b, and extracting 5 multi-fractal characteristics of the S1-Sm sections: singular index of maximum fluctuation alphamaxSingular index of minimum fluctuation alphaminThe width of the multi-fractal spectrum Delta alpha is alphamax-αminSingular index alpha corresponding to an extremum point of a multifractal spectrum0(fmax＝f(α0)，α0∈[αmin，αmax]) The fractal dimension difference delta f (alpha) of the probability subset of the multi-fractal spectrum is fmax)-f(αmin)；
e. Performing serial combination according to the 5 Alpha stable distribution characteristics and the 5 multi-fractal characteristics of the S1-Sm obtained by calculation in the steps c and d to obtain the combined characteristic set (Alpha, beta, gamma, h, Alpha) of the S1-Sm0，αmin，αmax，Δα，Δf)；
f. Taking the radial basis as a kernel function, performing dimensionality reduction fusion on the combined feature set in the step e by using a Kernel Principal Component Analysis (KPCA), and selecting a kernel principal element according to the variance cumulative contribution rate of more than or equal to 95% to obtain a new principal element fusion feature set;
g. taking the principal component fusion feature set obtained in the step f as an input sample, and optimizing two core parameters (a normalized parameter lambda and an inner core parameter sigma) of a least square support vector machine by utilizing a particle swarm optimization algorithm to establish a PSO-LSSVM model by using the obtained optimal parameters;
h. replacing the test speed changer with a speed changer to be tested, pasting a vibration acceleration sensor on the outer ring of one or more bearings to be tested in the speed changer to be tested, repeating the steps a to f, and bringing the principal component fusion feature sets of S1-Sm of the bearing to be tested, which are acquired by the vibration acceleration sensor or sensors in the step f, into the trained PSO-LSSVM model for state classification; and finishing the diagnosis.
The invention has the beneficial effects that: the vibration acceleration signal of the bearing is decomposed and reconstructed by an EMD method, so that the noise in the bearing signal of the automobile transmission can be effectively inhibited, and useful components in the signal are enhanced and highlighted; the signals subjected to EMD noise reduction are subjected to Alpha stable distribution and multi-fractal feature extraction respectively, and feature fusion is performed by utilizing kernel principal component analysis, so that the respective advantages of Alpha stable distribution and multi-fractal can be fully combined, the effectiveness of the features is maximized, the accuracy and efficiency of fault diagnosis are improved, a basic PSO-LSSVM model is established by relevant data collected in a test bed by a test gearbox with bearings of different fault types and different fault degrees, the relevant data collected in the test bed by the tested gearbox is brought into the established PSO-LSSVM model, and accordingly the position of the bearing with the fault in the tested gearbox and the fault degree of the bearing are obtained through analysis, the diagnosis efficiency is high, and the accuracy is high.
Drawings
FIG. 1 is a schematic diagram of a bearing test bed structure;
FIG. 2 is a schematic diagram of a bearing test stand control circuit;
FIG. 3 is a flow chart of EMD noise reduction for vibration acceleration signals;
FIG. 4 is a schematic diagram of ten features extracted by Alpha stable distribution parameter estimation and multi-fractal detrending fluctuation analysis on an acceleration signal with highlighted features after EMD noise reduction;
fig. 5 is a flowchart of the operation of quantitative diagnosis of bearing faults by using the bearing test stand.
Detailed Description
A bearing test bed as shown in fig. 1-2 comprises a base 1, a liftable three-coordinate electromagnetic vibration device 2 is arranged on the base 1, a truss type clamping tool 3 formed by splicing a plurality of channel steel or rectangular steel is arranged on the electromagnetic vibration device 2, and the position, the size and the vertical rigidity of a mounting bolt hole arranged at the corresponding position on the clamping tool 3 are matched with the mounting bolt hole of a test transmission 4; the structure of the test transmission 4 is consistent with that of the tested automobile transmission, and one or more fault bearings are arranged at a plurality of positions where the bearings are arranged in the test transmission;
the base 1 is connected with the electromagnetic vibration device 2 through an air spring 11 and a vertical vibration absorber 12, an air inlet and an air outlet of the air spring 11 are respectively connected with an air inlet control valve 13 and an air release valve 14, and the air inlet control valve 13 is connected with an electric air pump 15; before testing, the central processing unit 5 controls the air inlet control valve 13, the air release valve 14 and the electric air pump 15 to enable the vertical rigidity of the air spring 11 to be consistent with the vertical rigidity of the tested automobile suspension, and adjusts the damping coefficient of the vertical shock absorber 12 to enable the damping coefficient to be consistent with the vertical damping coefficient of the tested automobile suspension; the damping coefficient of the vertical shock absorber 12 can be manually adjusted, or an active vertical shock absorber 12 can be set and automatically adjusted by the central processing unit 5;
the input shaft of the test transmission 4 is connected with the output end of a torque loading device, the torque loading device comprises a motor 71, the output end of the motor 71 is sequentially connected with a torque sensor 72 and the input end of a fixed gear reduction gear 73, and the output end of the fixed gear reduction gear 73 is connected with the input shaft of the test transmission 4; a rotating speed sensor 74 is also arranged on the output end of the motor 71;
the output shaft of the test transmission 4 is connected with an inertia load device, the inertia load device comprises an auxiliary support 81 which is mutually independent from the base 1, a transmission gear set 82 is arranged on the auxiliary support 81, the transmission gear set 82 is composed of a pair of cylindrical straight gears or a pair of conical gears, a driving gear of the transmission gear set 82 is arranged on the output shaft of the test transmission 4, a driven gear of the transmission gear set 82 is connected with a rotating wheel 83 through a middle shaft, the rotating wheel 83 is arranged on the auxiliary support 81 through a bearing, a hydraulic brake caliper 84 which is adaptive to the size of the rotating wheel 83 is also arranged on the auxiliary support 81,
the outer ring of one or more faulty bearings installed in the test transmission 4 is pasted with a vibration acceleration sensor 41; the central processing unit 5 is respectively in communication connection with a hydraulic cylinder control valve of the hydraulic brake caliper 84, a rotating speed sensor 74, a torque sensor 72, the electric motor 71, a gear shifting controller 42 of the test transmission 4, a plurality of vibration acceleration sensors 41, an air inlet control valve 13, an air release valve 14, an electric air pump 15 and the electromagnetic vibration device 2;
as shown in fig. 5, the method of detecting the state of each bearing in the transmission of the automobile includes the steps of:
setting bearings at one or more positions in the test transmission 4 as fault bearings, and pasting a vibration acceleration sensor 41 on the outer ring of each fault bearing, wherein the fault bearings can be set with different fault positions and fault degrees, for example, the fault positions can be the outer ring, the inner ring, the roller and the retainer, and the fault degrees of the fault bearings can be respectively an early stage, a middle stage and a late stage;
then the following steps are carried out in sequence:
a. the central processing unit 5 controls the electromagnetic vibration device 2 to generate specific amplitude and vibration frequency, and the electromagnetic vibration device 2 can only apply vibration in the vertical direction and can also provide transverse and longitudinal vibration according to the requirement; meanwhile, the central processing unit 5 also controls the motor 71 and the gear shifting controller 42 to enable the output shaft of the test transmission 4 to output a specific rotating speed, and the rotating speed sensor 74 detects the output rotating speed of the motor 71 to ensure that the output rotating speed of the test transmission 4 is correct; the central processing unit 5 controls the brake calipers 84 to apply brake torque to the rotating wheel 83 while the output shaft of the test transmission 4 outputs a specific rotating speed, so that the output shaft of the test transmission 4 is subjected to a specific load torque; the torque sensor 72 detects the braking torque applied by the brake caliper 84 to ensure that the load torque applied to the test transmission 4 is correct; the actual line can be led into the central processing unit 5, the central processing unit 5 respectively controls the electromagnetic vibration device 2, the motor 71 and the brake caliper 84 according to the actual road condition, so that the environmental vibration parameters, the speed of the output shaft of the test transmission 4 and the load torque of the output shaft are matched with the actual condition, and the test result is more accurate;
the vibration acceleration sensor 41 collects a vibration acceleration signal sample of the fault bearing; taking the collected sample as a training sample, and performing EMD self-adaptive decomposition on a vibration acceleration signal x (t) in the training sample, wherein the decomposition method comprises the following steps:
in the above formula, n is the number of decomposed IMF components; cjRepresents the jth IMF component, j 1, 2, 3.., n; r isnIs the residual component;
b. decomposing the mixture to obtain n CjAfter the components, each C is calculated separatelyj(j ═ 1, 2, 3.., n) kurtosis value, two of C with the highest kurtosis value and the second highest kurtosis value are selectediPerforming linear superposition to obtain an acceleration signal of the characteristic highlight subjected to EMD noise reduction, then averagely dividing the obtained acceleration signal of the characteristic highlight into m sections according to the time length, and recording the signals of different time length sections as S1-Sm;
c. respectively estimating Alpha stable distribution parameters of the S1-Sm sections in the step b and calculating probability density functions of the sections, and extracting 5 Alpha stable distribution characteristics of characteristic index Alpha (Alpha is more than 0 and less than or equal to 2), symmetric parameter beta (-beta is more than or equal to 1), dispersion coefficient gamma (gamma is more than 0), position parameter (-infinity is more than or equal to infinity) and extremum h (h is more than 0) of the probability density functions;
d. respectively carrying out multi-fractal detrending waves on the S1-Sm sections in the step bAnd (3) performing dynamic analysis, and extracting 5 multi-fractal characteristics of S1-Sm respectively: singular index of maximum fluctuation alphamaxSingular index of minimum fluctuation alphaminThe width of the multi-fractal spectrum Delta alpha is alphamax-αminSingular index alpha corresponding to an extremum point of a multifractal spectrum0(fmax＝f(α0)，α0∈[αmin，αmax]) The fractal dimension difference delta f (alpha) of the probability subset of the multi-fractal spectrum is fmax)-f(αmin)；
e. Performing serial combination according to 5 Alpha stable distribution characteristics and 5 multi-fractal characteristics of S1-Sm obtained by calculation in the steps c and d to obtain a combined characteristic set (Alpha, beta, gamma, h, Alpha) of each section of S1-Sm0，αmin，αmax，Δα，Δf)；
f. Taking the radial basis as a kernel function, performing dimensionality reduction fusion on the combined feature set in the step e by using a Kernel Principal Component Analysis (KPCA), and selecting a kernel principal element according to the variance cumulative contribution rate of more than or equal to 95% to obtain a new principal element fusion feature set;
g. taking the principal component fusion feature set obtained in the step f as an input sample, and optimizing two core parameters (a normalized parameter lambda and an inner core parameter sigma) of a least square support vector machine by utilizing a particle swarm optimization algorithm to establish a PSO-LSSVM model by using the obtained optimal parameters;
h. replacing the test speed changer 4 with a speed changer to be tested, pasting a vibration acceleration sensor 41 on the outer ring of one or more bearings to be tested which need to be tested and have the same position as the fault bearing in the test speed changer 4 in the speed changer to be tested, then repeating the steps a to f, and bringing the principal element fusion characteristic set of each section S1-Sm of the bearing to be tested, which is acquired by one or more vibration acceleration sensors 41 in the step f, into the trained PSO-LSSVM model for state classification; and finishing the diagnosis.
Examples of the above diagnostic method are as follows: replacing bearings at a specific position in the test transmission 4 with faulty bearings having outer ring faults (3 damage degrees divided into early, middle and late stages), inner ring faults (early, middle and late stages), roller faults (early, middle and late stages) and retainer faults (early, middle and late stages) for multiple times, wherein only one faulty bearing is arranged in the test transmission 4 after each replacement; respectively installing a test transmission 4 on a test bed, carrying out steps a to b, averagely dividing an acceleration signal obtained by a vibration acceleration sensor 41 adhered to a fault bearing into 5 sections according to a time period, wherein the total duration of the acceleration signal is 0.3s, the duration of each section of signal is 0.06s, and carrying out steps c to f respectively aiming at the 5 sections of signals;
taking the principal component fusion feature set calculated in the step f as an input sample, and optimizing two core parameters (a normalized parameter lambda and a kernel parameter sigma) of a least square support vector machine by utilizing a particle swarm optimization algorithm to obtain an optimal parameter to establish a PSO-LSSVM model;
then placing a tested bearing with unknown faults at the same position in the tested speed changer, mounting the tested speed changer on a test bed, and repeating the steps a to b, wherein the EMD noise reduction effect of the step b is shown in FIG. 3;
averagely dividing the acquired EMD denoised acceleration signal into 5 sections according to time periods, wherein the measurement time of the acceleration signal is 0.3S, the time length of each averagely distributed section of signal is 0.06S, signals of sections with different time lengths are marked as S1-S5, and steps c to f are carried out on the 5 sections of signals; the process of performing Alpha stable distribution parameter estimation on the S1-S5 with characteristic highlighting after EMD noise reduction and performing multi-fractal detrending fluctuation analysis on the S1-S5 with characteristic highlighting after EMD noise reduction is shown in FIG. 4;
wherein the combined feature set calculated in step e from S1-S5 is shown in Table 1:
TABLE 1 Combined feature set parameters of S1-S5
In the step f, the combined feature set of S1-S5 is fused by using kernel principal component analysis, kernel principal elements are selected according to the variance cumulative contribution rate of greater than or equal to 95%, and the obtained kernel principal element fusion features are shown in Table 2:
TABLE 2 Kernel-member fusion characteristics of S1-S5
Then, bringing the pivot fusion feature set of S1-S5 calculated in the step f into a trained PSO-LSSVM model to classify the state of the model; namely, the five-dimensional kernel principal component fusion features shown in table 2 are input into the established PSO-LSSVM classifier for classification, and the classification result is shown in table 3.
TABLE 3 results of the classification
According to the classification results shown in table 3, 5 sections of signals from S1 to S5 indicate that the failure mode of the tested bearing in the tested gearbox is that the outer ring of the bearing is damaged early, according to the above process, the bearing with inner ring failure, roller failure or retainer failure is put into the tested gearbox 4, and the steps from a to g are repeated, so that the state of each bearing in the tested gearbox can be accurately detected by using the trained PSO-LSSVM model.
Claims (1)
1. The method for detecting the state of each bearing in the automobile transmission comprises a test transmission (4) and a tested transmission, wherein the test transmission (4) and the tested transmission can be installed on a bearing test bed, the bearing test bed comprises a base (1), a liftable three-coordinate type electromagnetic vibration device (2) is arranged on the base (1), a truss type clamping tool (3) formed by splicing a plurality of channel steel or rectangular steel is arranged on the electromagnetic vibration device (2), the position, the size and the vertical rigidity of an installation bolt hole arranged at a corresponding position on the clamping tool (3) are matched with the installation bolt hole of the test transmission (4), the structure of the test transmission (4) is consistent with that of the tested automobile transmission, and one or more fault bearings are arranged at a plurality of positions where the bearings are arranged in the test transmission;
the base (1) is connected with the electromagnetic vibration device (2) through an air spring (11) and a vertical vibration absorber (12), an air inlet and an air outlet of the air spring (11) are respectively connected with an air inlet control valve (13) and an air release valve (14), the air inlet control valve (13) is connected with an electric air pump (15), the vertical rigidity of the air spring (11) is consistent with the vertical rigidity of a tested automobile suspension, and the damping coefficient of the vertical vibration absorber (12) is consistent with the vertical damping coefficient of the tested automobile suspension;
the input shaft of the test transmission (4) is connected with the output end of a torque loading device, the torque loading device comprises a motor (71), the output end of the motor (71) is sequentially connected with a torque sensor (72) and the input end of a fixed-gear-ratio speed reducer (73), and the output end of the fixed-gear-ratio speed reducer (73) is connected with the input shaft of the test transmission (4); a rotating speed sensor (74) is also arranged on the output end of the motor (71);
the output shaft of experimental derailleur (4) is connected with the inertial load device, the inertial load device includes auxiliary support (81) with base (1) mutual independence sets up, install transmission gear group (82) on auxiliary support (81), transmission gear group (82) comprise a pair of cylinder spur gear or a pair of conical gear, the driving gear of transmission gear group (82) is installed on the output shaft of experimental derailleur (4), the driven gear of transmission gear group (82) passes through the jackshaft and is connected with swiveling wheel (83), swiveling wheel (83) pass through the bearing and install on auxiliary support (81), still set up hydraulic braking calliper (84) that suits with swiveling wheel (83) size on auxiliary support (81),
one or more outer ring sticking vibration acceleration sensors (41) with fault bearings are arranged in the test transmission (4); the central processing unit (5) is respectively in communication connection with a hydraulic cylinder control valve, a rotating speed sensor (74), a torque sensor (72), a motor (71), a gear shifting controller (42) of the test transmission (4), a plurality of vibration acceleration sensors (41), an air inlet control valve (13), an air release valve (14), an electric air pump (15) and an electromagnetic vibration device (2) of the hydraulic brake caliper (84);
the method is characterized in that: the method for detecting the state of each bearing in the automobile transmission comprises the following steps:
setting bearings at one or more positions in a test transmission (4) as fault bearings, pasting a vibration acceleration sensor (41) on an outer ring of each fault bearing, setting different fault positions and fault degrees of the fault bearings, wherein the fault positions are respectively an outer ring, an inner ring, a roller and a retainer, and the fault degrees of the fault bearings are respectively an early stage, a middle stage and a late stage; then the following steps are carried out in sequence:
a. the central processing unit (5) controls the electromagnetic vibration device (2) to generate specific amplitude and vibration frequency; meanwhile, the central processing unit (5) controls the motor (71) and the gear shifting controller (42) to enable an output shaft of the test transmission (4) to output a specific rotating speed; when the output shaft of the test transmission (4) outputs a specific rotating speed, the central processing unit (5) controls the brake calipers (84) to apply brake torque to the rotating wheel (83), so that the output shaft of the test transmission (4) is subjected to specific load torque; a vibration acceleration sensor (41) collects a vibration acceleration signal sample of a fault bearing; taking the collected sample as a training sample, and performing EMD self-adaptive decomposition on a vibration acceleration signal x (t) in the training sample, wherein the decomposition method comprises the following steps:
in the above formula, n is the number of decomposed IMF components; cjRepresents the jth IMF component, j ═ 1, 2, 3 …, n; r isnIs the residual component;
b. decomposing the mixture to obtain n CjAfter the components, each C is calculated separatelyj1, 2, 3 …, n; selecting two C with the maximum kurtosis value and the second-order kurtosis valuejPerforming linear superposition to obtain an acceleration signal of the characteristic highlight subjected to EMD noise reduction, then averagely dividing the obtained acceleration signal of the characteristic highlight into m sections according to the time length, and recording the signals of different time length sections as S1-Sm;
c. respectively estimating Alpha stable distribution parameters of each section S1-Sm in the step b, calculating a probability density function of each section S1-Sm, extracting characteristic indexes Alpha, wherein Alpha is more than 0 and less than or equal to 2, symmetric parameters beta, wherein beta is more than or equal to-1 and less than or equal to 1, dispersion coefficients gamma, wherein gamma is more than 0, position parameters, wherein infinity is more than or equal to infinity and more than or equal to infinity, and an extremum h of the probability density function, wherein h is more than 0, and 5 Alpha stable distribution characteristics are obtained;
d. and (c) respectively carrying out multi-fractal detrending fluctuation analysis on the S1-Sm sections in the step b, and extracting 5 multi-fractal characteristics of the S1-Sm sections: singular index of maximum fluctuation alphamaxSingular index of minimum fluctuation alphaminThe width of the multi-fractal spectrum Delta alpha is alphamax-αminSingular index alpha corresponding to an extremum point of a multifractal spectrum0，fmax＝f(α0)，α0∈[αmin，αmax]The fractal dimension difference delta f (alpha) of the probability subset of the multi-fractal spectrum is fmax)-f(αmin)；
e. Performing serial combination according to the 5 Alpha stable distribution characteristics and the 5 multi-fractal characteristics of the S1-Sm obtained by calculation in the steps c and d to obtain the combined characteristic set (Alpha, beta, gamma, h, Alpha) of the S1-Sm0，αmin，αmax，Δα，Δf)；
f. Taking the radial basis as a kernel function, performing dimensionality reduction fusion on the combined feature set in the step e by using a Kernel Principal Component Analysis (KPCA), and selecting a kernel principal element according to the variance cumulative contribution rate of more than or equal to 95% to obtain a new principal element fusion feature set;
g. taking the principal component fusion feature set obtained in the step f as an input sample, and optimizing two core parameter normalization parameters lambda and an inner core parameter sigma of a least square support vector machine by utilizing a particle swarm optimization algorithm to obtain an optimal parameter to establish a PSO-LSSVM model;
h. replacing the test speed changer (4) with a speed changer to be tested, pasting a vibration acceleration sensor (41) on the outer ring of one or more bearings to be tested in the speed changer to be tested, wherein the positions of the one or more bearings to be tested are the same as the position of a fault bearing, repeating the steps a to f, and bringing the principal component fusion feature sets of S1-Sm of the bearing to be tested, which are acquired by the vibration acceleration sensor (41) in the step f, into the trained PSO-LSSVM model for state classification; and finishing the detection.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811588388.6A CN109708890B (en) | 2017-08-09 | 2017-08-09 | Method for detecting the condition of bearings in a motor vehicle transmission |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811588388.6A CN109708890B (en) | 2017-08-09 | 2017-08-09 | Method for detecting the condition of bearings in a motor vehicle transmission |
| CN201710678357.9A CN107271187B (en) | 2017-08-09 | 2017-08-09 | The method that a kind of pair of automobile speed variator bearing failure carries out quantitative Diagnosis |
Related Parent Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710678357.9A Division CN107271187B (en) | 2017-08-09 | 2017-08-09 | The method that a kind of pair of automobile speed variator bearing failure carries out quantitative Diagnosis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109708890A CN109708890A (en) | 2019-05-03 |
| CN109708890B true CN109708890B (en) | 2020-10-30 |
Family
ID=60077043
Family Applications (3)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811588388.6A Active CN109708890B (en) | 2017-08-09 | 2017-08-09 | Method for detecting the condition of bearings in a motor vehicle transmission |
| CN201710678357.9A Active CN107271187B (en) | 2017-08-09 | 2017-08-09 | The method that a kind of pair of automobile speed variator bearing failure carries out quantitative Diagnosis |
| CN201811597766.7A Active CN109668734B (en) | 2017-08-09 | 2017-08-09 | Quantitative diagnosis method for parts in automobile transmission bearing |
Family Applications After (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710678357.9A Active CN107271187B (en) | 2017-08-09 | 2017-08-09 | The method that a kind of pair of automobile speed variator bearing failure carries out quantitative Diagnosis |
| CN201811597766.7A Active CN109668734B (en) | 2017-08-09 | 2017-08-09 | Quantitative diagnosis method for parts in automobile transmission bearing |
Country Status (1)
| Country | Link |
|---|---|
| CN (3) | CN109708890B (en) |
Families Citing this family (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108105081B (en) * | 2018-02-05 | 2023-10-27 | 中国石油化工股份有限公司 | Water injection pump unit fault diagnosis method and device |
| CN108498106B (en) * | 2018-02-08 | 2021-10-29 | 陕西师范大学 | CNV electroencephalogram lie detection method based on multi-fractal elimination trend fluctuation analysis |
| CN108460429B (en) * | 2018-05-03 | 2019-03-05 | 武汉理工大学 | Application method of the multiple regression LSSVM model in the diagnosis of mechanical concurrent fault |
| CN109374293B (en) * | 2018-10-29 | 2020-07-24 | 珠海市华星装备信息科技有限公司 | Gear fault diagnosis method |
| CN109827777B (en) * | 2019-04-01 | 2020-12-18 | 哈尔滨理工大学 | Rolling bearing fault prediction method based on partial least square method extreme learning machine |
| CN110159428B (en) * | 2019-06-05 | 2024-03-12 | 西华大学 | Device and method for diagnosing carbon deposit in engine cylinder |
| CN110046476B (en) * | 2019-06-05 | 2020-10-16 | 厦门大学 | Ternary binary fractal wavelet sparse diagnosis method for rolling bearing faults |
| CN111120348A (en) * | 2019-12-25 | 2020-05-08 | 中国石化销售股份有限公司华南分公司 | Centrifugal pump fault early warning method based on support vector machine probability density estimation |
| CN111767891A (en) * | 2020-07-09 | 2020-10-13 | 南京工程学院 | Fault diagnosis method for rolling bearing |
| CN112037500A (en) * | 2020-07-21 | 2020-12-04 | 江苏国茂减速机股份有限公司 | Speed reducer capable of automatically judging fault, elevator and fault automatic judging method |
| CN112749453B (en) * | 2020-12-16 | 2023-10-13 | 安徽三禾一信息科技有限公司 | Complex equipment residual service life prediction method based on improved SVR |
| CN112597958B (en) * | 2020-12-29 | 2023-04-07 | 哈工大机器人（合肥）国际创新研究院 | Automatic identification method and system for rolling bearing fault |
| CN113804470B (en) * | 2021-04-14 | 2023-12-01 | 山东省计算中心（国家超级计算济南中心） | Fault detection feedback method for plug seedling production line |
| CN113246175B (en) * | 2021-05-07 | 2024-05-03 | 配天机器人技术有限公司 | Robot terminal vibration classification method, robot and storage medium |
| CN113281035B (en) * | 2021-05-31 | 2022-12-09 | 重庆长安汽车股份有限公司 | Method for testing locking abnormal sound of automatic transmission |
| CN113408371A (en) * | 2021-06-01 | 2021-09-17 | 武汉理工大学 | Early fault diagnosis method and device |
| CN113628175B (en) * | 2021-07-22 | 2024-02-20 | 上海交通大学 | Image quality score distribution prediction method, system, terminal and medium |
| CN115266905B (en) * | 2022-08-02 | 2024-05-07 | 山东申正汽车拆解有限公司 | Crack detection method and equipment for recycling and remanufacturing engine crankshaft of heavy-duty commercial vehicle |
| CN117330941B (en) * | 2023-11-29 | 2024-02-02 | 珠海智锐科技有限公司 | Performance test method and related device for new energy BMS printed circuit board |
Family Cites Families (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE10333410B4 (en) * | 2003-07-15 | 2017-03-23 | Minebea Co., Ltd. | Method and device for determining the natural frequencies of a bearing system with a mounted shaft |
| CN102564756B (en) * | 2010-12-31 | 2015-06-03 | 中国科学院沈阳自动化研究所 | Automobile transmission vibration analysis testing method and device |
| CN102539151B (en) * | 2012-01-18 | 2014-05-14 | 北京工业大学 | Intelligent online quality detection method for automobile transmission |
| JP6133112B2 (en) * | 2013-04-11 | 2017-05-24 | Ntn株式会社 | Rolling bearing diagnostic device and rolling bearing diagnostic method |
| US9841352B2 (en) * | 2014-06-19 | 2017-12-12 | United Technologies Corporation | System and method for monitoring gear and bearing health |
| CN104697787B (en) * | 2015-03-20 | 2019-03-26 | 山东大学 | A kind of gearbox test-bed and its detection method based on multi-information fusion |
| CN105172511B (en) * | 2015-07-15 | 2017-09-26 | 西南交通大学 | A kind of shock absorber of automobile suspension control system and method |
| JP6714806B2 (en) * | 2015-08-06 | 2020-07-01 | 日本精工株式会社 | Status monitoring device and status monitoring method |
| CN106441861A (en) * | 2015-08-10 | 2017-02-22 | 中国石油天然气股份有限公司 | Automatic transmission oil transmission efficiency detection device and automatic transmission oil transmission efficiency detection method |
| CN106092310A (en) * | 2016-04-21 | 2016-11-09 | 重庆理工大学 | A kind of automotive transmission vibration noise off-line test method |
| CN106017879B (en) * | 2016-05-18 | 2018-07-03 | 河北工业大学 | Omnipotent breaker mechanical failure diagnostic method based on acoustic signal Fusion Features |
| CN105784361B (en) * | 2016-05-30 | 2018-01-16 | 吉林大学 | Dynamic power machine closed loop tilting polygon exciting gear box test table |
| CN106053072A (en) * | 2016-07-20 | 2016-10-26 | 西南交通大学 | High-speed train axle box bearing test bench based on vibration rolling |
| CN206074255U (en) * | 2016-07-20 | 2017-04-05 | 西南交通大学 | A kind of bullet train axle box bearing testing stand based on rolling vibration |
| CN106840651A (en) * | 2017-01-16 | 2017-06-13 | 广州汽车集团股份有限公司 | Speed changer heat examination platform |
-
2017
- 2017-08-09 CN CN201811588388.6A patent/CN109708890B/en active Active
- 2017-08-09 CN CN201710678357.9A patent/CN107271187B/en active Active
- 2017-08-09 CN CN201811597766.7A patent/CN109668734B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN109708890A (en) | 2019-05-03 |
| CN109668734B (en) | 2020-09-04 |
| CN107271187A (en) | 2017-10-20 |
| CN109668734A (en) | 2019-04-23 |
| CN107271187B (en) | 2019-01-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109708890B (en) | Method for detecting the condition of bearings in a motor vehicle transmission | |
| CN110866365B (en) | Mechanical equipment intelligent fault diagnosis method based on partial migration convolutional network | |
| CN109822399B (en) | Numerical control machine tool cutter wear state prediction method based on parallel deep neural network | |
| CN103323244B (en) | Bearing accuracy performance testing device and method of testing and determine the method for bearing accuracy initial failure part | |
| CN111521406B (en) | High-speed wind noise separation method for passenger car road test | |
| Sevilla-Camacho et al. | FPGA-based reconfigurable system for tool condition monitoring in high-speed machining process | |
| CN112991577B (en) | Railway vehicle wheel polygon state diagnostic system | |
| CN207066742U (en) | A kind of automobile speed variator bearing testing stand | |
| CN114755017B (en) | Variable-speed bearing fault diagnosis method of cross-domain data driving unsupervised field shared network | |
| CN112525534A (en) | Bearing and gear damage quantitative evaluation method based on acoustic emission technology | |
| CN114871850B (en) | Tool wear state assessment method based on vibration signals and BP neural network | |
| CN104931262A (en) | Bearing fault diagnosis device used in cooperation with wheel set bearing running-in machine and diagnosis method thereof | |
| CN116793682A (en) | Bearing fault diagnosis method based on iCORAL-MMD and anti-migration learning | |
| CN113933074A (en) | Suspension assembly road simulation test method based on standardized load spectrum | |
| CN111337767A (en) | Resonant wave reducer fault analysis method | |
| CN112766200A (en) | Planetary gear box small sample fault diagnosis method based on one-dimensional convolutional neural network | |
| CN115496107A (en) | Aero-engine main shaft bearing full life cycle analysis method based on digital twinning | |
| CN112441254B (en) | Performance detection method and device of engine lubricating oil system, storage medium and terminal | |
| CN112161805A (en) | Bearing tiny fault diagnosis method based on time series scale analysis and CNN | |
| Unal et al. | Fault diagnosis of rolling bearing based on feature extraction and neural network algorithm | |
| CN111351655A (en) | Method for diagnosing gear fault in locomotive gear box | |
| CN111337255A (en) | Rolling bearing peeling position determination method | |
| GE et al. | Fault Detection of Spindle Rolling Bearings in CNC Machine Tools Based on Odorless Kalman Filtering Algorithm | |
| CN114894138B (en) | Method, system, equipment and storage medium for detecting vertical displacement difference of automobile wheel | |
| Saimurugan et al. | On-road testing of a vehicle for gearbox fault detection using vibration signals |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20211116 Address after: 224006 No. 6, Yuxing Road, Zhongxing Street, Xuefu Town, Yandu District, Yancheng City, Jiangsu Province (P) Patentee after: Fuller transmission equipment Yancheng Co., Ltd Address before: 610039 Xihua University, No. 999 Jinzhou Road, Jinniu District, Chengdu, Sichuan Patentee before: Xihua University |