CN117390519B - Wheel hub motor fault condition prediction method - Google Patents
Wheel hub motor fault condition prediction method Download PDFInfo
- Publication number
- CN117390519B CN117390519B CN202311656884.1A CN202311656884A CN117390519B CN 117390519 B CN117390519 B CN 117390519B CN 202311656884 A CN202311656884 A CN 202311656884A CN 117390519 B CN117390519 B CN 117390519B
- Authority
- CN
- China
- Prior art keywords
- hub motor
- fault
- test
- wheel
- data
- 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
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000012360 testing method Methods 0.000 claims abstract description 130
- 239000011159 matrix material Substances 0.000 claims abstract description 74
- 238000004458 analytical method Methods 0.000 claims abstract description 66
- 238000013528 artificial neural network Methods 0.000 claims abstract description 33
- 238000001228 spectrum Methods 0.000 claims description 20
- 230000008569 process Effects 0.000 claims description 7
- 230000007774 longterm Effects 0.000 abstract description 7
- 230000005284 excitation Effects 0.000 description 17
- 230000001133 acceleration Effects 0.000 description 12
- 230000004044 response Effects 0.000 description 11
- 238000012549 training Methods 0.000 description 11
- 230000008859 change Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 8
- 239000000725 suspension Substances 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 7
- 238000010276 construction Methods 0.000 description 4
- 238000006073 displacement reaction Methods 0.000 description 4
- 238000012937 correction Methods 0.000 description 3
- 238000003062 neural network model Methods 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000011065 in-situ storage Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000005316 response function Methods 0.000 description 2
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000017525 heat dissipation Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035699 permeability Effects 0.000 description 1
- 238000007789 sealing Methods 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 230000000087 stabilizing effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/34—Testing dynamo-electric machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0499—Feedforward networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention relates to the technical field of hub motor system and assembly test, and discloses a method for predicting the failure condition of a hub motor, which comprises the following steps: basic data required by a wheel hub motor fault test are collected; the wheel hub motor fault test comprises a wheel hub motor whole vehicle fault test or a wheel hub motor angle module fault test; developing a wheel hub motor fault test and recording fault analysis data of a wheel hub motor angle module; determining a weight matrix of the basic data and the fault analysis data according to the basic data and the fault analysis data; and inputting the basic data and the weight matrix of the hub motor to be tested into a neural network prediction model, and predicting the fault condition of the hub motor to be tested. The use requirement of the wheel hub motor on long-term stable service under the running condition of the whole vehicle and the independent angle module can be rapidly and effectively verified.
Description
Technical Field
The invention relates to the technical field of hub motor system and assembly test, in particular to a method for predicting the failure condition of a hub motor.
Background
With the continuous improvement of the market permeability of new energy automobiles and the technical development of wire control chassis and skateboard chassis, the continuous breakthrough of the key technology of the hub motor is driven, and the application field of the hub motor automobiles is also widened continuously. The wheel hub motor angle module for the automobile integrates a wheel hub motor, a suspension system, a steering system and a braking system, has the advantages of high-efficiency driving mode, compact structural design, flexible control mode and the like, but also has the disadvantages of long product verification period, high requirements on the safety, reliability and durability of the system and the like.
Compared with the traditional centralized control motor, the hub motor is arranged in the wheel, so that the running environment required by an automobile product is greatly changed, and the hub motor is required to run for a long time under more severe environmental conditions. Hub motors located under the springs of the vehicle suspension bear greater impact and vibration than centralized control motors located on the springs due to potholes and other road surface imperfections. Hub motors are also more susceptible to environmental factors such as dirt and water, total flooding, mud, etc. In addition, hub motors are subject to more complex mechanical loads such as side-to-side cornering loads, cable motion due to suspension operation, and potential roadside impact.
Therefore, a method for predicting the failure condition of the hub motor is needed to rapidly and effectively verify the long-term stable service use requirement of the hub motor under the running conditions of the whole vehicle and the independent angle modules.
Disclosure of Invention
In order to solve the technical problems, the invention provides a method for predicting the failure condition of a hub motor, which is used for rapidly and effectively verifying the use requirement of the hub motor for long-term stable service under the running conditions of a whole vehicle and an independent angle module.
The invention provides a method for predicting the failure condition of a hub motor, which comprises the following steps:
s1, acquiring basic data required by a wheel hub motor fault test; the wheel hub motor fault test comprises a wheel hub motor whole vehicle fault test or a wheel hub motor angle module fault test; when the wheel hub motor fault test is a wheel hub motor complete vehicle fault test, the basic data comprise vehicle basic data, and when the wheel hub motor fault test is a wheel hub motor angle module fault test, the basic data comprise angle module basic data;
s2, developing a wheel hub motor fault test, and recording fault analysis data of a wheel hub motor angle module;
s3, determining a weight matrix of the basic data and the fault analysis data according to the basic data and the fault analysis data;
and S4, inputting basic data and a weight matrix of the hub motor to be tested into a neural network prediction model, and predicting the fault condition of the hub motor to be tested.
Further, when the in-wheel motor failure test is an in-wheel motor complete vehicle failure test, after S1, the method further includes:
collecting load spectrum data of a first hub motor angle module required by a whole-vehicle fault test of the hub motor, and extracting a first load index according to the load spectrum data of the first hub motor angle module;
and calculating driving parameters for the complete vehicle fault test of the hub motor according to the load spectrum data of the first hub motor angle module.
Further, S2, developing a failure test of the hub motor, and recording failure analysis data of the hub motor corner module includes:
s21', carrying out a complete vehicle fault test of the hub motor according to the driving parameters, and collecting load spectrum data of a second hub motor angle module in the test process;
s22', extracting a second load index according to load spectrum data of a second hub motor angle module;
and S23', after the test is finished, recording fault analysis data of the hub motor angle module.
Further, S3, determining the weight matrix of the base data and the failure analysis data according to the base data and the failure analysis data includes:
s31', constructing a first weight matrix W according to the basic data of the vehicle and the first load index 1 ；
S32', constructing a second weight matrix W according to the second load index and the fault analysis data 2 ；
S33' according to the first weight matrix W 1 And a second weight matrix W 2 A third weight matrix W for deriving vehicle base data and fault analysis data 3 。
Further, S4, inputting the basic data and the weight matrix of the hub motor to be tested into the neural network prediction model, and predicting the fault condition of the hub motor to be tested includes:
taking the basic data of the vehicle as input and taking a third weight matrix W 3 And inputting the initial weight into a neural network prediction model to predict the fault condition of the hub motor to be tested.
Further, when the in-wheel motor failure test is an in-wheel motor angle module failure test, after S1, it further includes:
determining a standard working condition of a wheel hub motor angle module fault test; the standard conditions include a start condition, a drive condition, a speed change condition, a steering condition, and a braking condition.
Further, S2, developing a failure test of the hub motor, and recording failure analysis data of the hub motor corner module includes:
and carrying out a wheel hub motor angle module fault test according to the standard working condition, and recording fault analysis data of the wheel hub motor angle module.
Further, S3, determining the weight matrix of the base data and the failure analysis data according to the base data and the failure analysis data includes:
constructing a fourth weight matrix W according to the corner module basic data and the fault analysis data 4 。
Further, S4, inputting the basic data and the weight matrix of the hub motor to be tested into the neural network prediction model, and predicting the fault condition of the hub motor to be tested includes:
taking the corner module basic data as input and taking a fourth weight matrix W 4 And inputting the initial weight into a neural network prediction model to predict the fault condition of the hub motor to be tested.
The embodiment of the invention has the following technical effects:
the method has the advantages that the basic data required by the wheel hub motor fault test and the fault analysis data of the wheel hub motor angle module are collected, the weight matrix of the basic data and the fault analysis data is determined, the fault condition of the wheel hub motor to be tested is obtained based on the neural network prediction model, the neural network prediction model is input according to different input parameters and different initialization weights under the condition that only the basic data of a vehicle or the basic data of the angle module are known, the fault condition of the wheel hub motor is predicted, the compatibility and universality of the neural network prediction model are improved, the fault condition of the wheel hub motor is considered more comprehensively, and the use requirement of the wheel hub motor for long-term stable service under the running condition of the whole vehicle and the single angle module can be verified rapidly and effectively.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of a method for predicting a failure condition of an in-wheel motor according to an embodiment of the present invention;
FIG. 2 is a flow chart of a complete vehicle failure test of a hub motor provided by an embodiment of the invention;
FIG. 3 is a logic diagram of a prediction method of a neural network prediction model according to an embodiment of the present invention;
FIG. 4 is a flow chart of a failure test of an in-wheel motor corner module provided by an embodiment of the invention;
FIG. 5 is a schematic diagram of a driving condition of an in-wheel motor angle module according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of a variable speed condition of an in-wheel motor angle module according to an embodiment of the present invention;
fig. 7 is a schematic diagram of a continuous acceleration and deceleration condition of an in-wheel motor angle module according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below. It will be apparent that the described embodiments are only some, but not all, embodiments of the invention. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the invention, are within the scope of the invention.
Fig. 1 is a flowchart of a method for predicting a failure condition of an in-wheel motor according to an embodiment of the present invention. Referring to fig. 1, the method specifically includes:
s1, basic data required by a wheel hub motor fault test are collected.
Specifically, the wheel hub motor fault test comprises a wheel hub motor whole vehicle fault test or a wheel hub motor angle module fault test; when the wheel hub motor fault test is a wheel hub motor whole vehicle fault test, the basic data comprise vehicle basic data, and when the wheel hub motor fault test is a wheel hub motor angle module fault test, the basic data comprise angle module basic data.
S2, developing a wheel hub motor fault test, and recording fault analysis data of the wheel hub motor corner module.
Specifically, the test duration of the failure test of the hub motor is the duration of durability specified by the technical requirements of the hub motor angle module. The failure analysis data of the wheel hub motor angle module comprise whether the frame structure of the wheel hub motor angle module is complete, whether each function is complete, whether a moving cable is complete, the fatigue damage condition of key parts, whether an electronic and electric element and a sensor are good and the like, and the time and the driving mileage of various failures are recorded for the first time. By way of example, data such as the operating time when the corner module fails, such as breaks, deformations or wear, the corresponding mileage, etc.
Illustratively, as shown in table 1, a fault classification statistics table is presented for recording fault analysis data for an in-wheel motor corner module:
TABLE 1 failure classification statistics table
Wherein, the first, second, third and fourth faults in table 1 represent fatal, severe, general and slight faults, respectively.
And S3, determining a weight matrix of the basic data and the fault analysis data according to the basic data and the fault analysis data.
Specifically, when the wheel hub motor fault test is a wheel hub motor complete vehicle fault test, determining a weight matrix of vehicle basic data and fault analysis data so as to determine the relationship between the vehicle basic data and the fault analysis data; when the wheel hub motor fault test is a wheel hub motor angle module fault test, determining a weight matrix of angle module basic data and fault analysis data so as to determine the relationship between the angle module basic data and the fault analysis data.
And S4, inputting basic data and a weight matrix of the hub motor to be tested into a neural network prediction model, and predicting the fault condition of the hub motor to be tested.
Specifically, basic data of the hub motor to be tested is used as input, a corresponding weight matrix is used as initial weight and is input into a neural network prediction model, and the fault condition of the hub motor to be tested is predicted to obtain fault analysis data of the hub motor to be tested under the whole vehicle condition or under the single angle module condition.
According to the embodiment of the invention, the basic data required by the wheel hub motor fault test and the fault analysis data of the wheel hub motor angle module are collected, the weight matrix of the basic data and the fault analysis data is determined, the fault condition of the wheel hub motor to be tested is obtained based on the neural network prediction model, the neural network prediction model can be input according to different input parameters and different initialization weights under the condition that only the basic data of a vehicle or the basic data of the angle module are known, the fault condition of the wheel hub motor is predicted, the compatibility and universality of the neural network prediction model are improved, the fault condition of the wheel hub motor is considered more comprehensively, and the use requirement of the wheel hub motor for long-term stable service under the running condition of the whole vehicle and the single angle module can be verified rapidly and effectively.
Fig. 2 is a flowchart of a complete vehicle failure test of a hub motor, referring to fig. 2, when the complete vehicle failure test of the hub motor is a complete vehicle failure test of the hub motor, the method includes the following steps:
s1', collecting vehicle basic data required by a wheel hub motor fault test.
Specifically, the vehicle base data may include vehicle mass, centroid position, tire model, tire pressure, angular module stiffness, angular module mass, bar system length, and the like.
Further, before the whole wheel hub motor fault test is carried out, load spectrum data of a first wheel hub motor angle module required by the whole wheel hub motor fault test are collected, and a first load index is extracted according to the load spectrum data of the first wheel hub motor angle module. The load spectrum data can comprise six component force data of the wheel center of the wheel hub motor angle module, acceleration data of the wheel center, displacement data of a suspension, stress data of a rod system and the like. The six-component force data of the wheel center are force and torque of the wheel center in the X, Y, Z triaxial directions; the wheel center acceleration data is acceleration of the wheel center in the X, Y, Z triaxial directions; the suspension displacement data is the position change of the suspension in the Z-axis direction; the stress data of the rod system is the stress condition in the axial direction of the connecting rod/the pull rod/the upper swing arm and the lower swing arm. The load index can comprise indexes such as peak value, peak value frequency and change rate of six component force of the wheel center, peak value frequency and acceleration change rate of the acceleration of the wheel center, peak value frequency and displacement change rate of the suspension displacement, peak value frequency and change rate of stress of the rod system and the like.
Further, driving parameters for the complete vehicle fault test of the hub motor are calculated according to the load spectrum data of the first hub motor angle module.
Specifically, a white noise signal is generated by setting frequency information and amplitude information, the white noise signal is used as an excitation signal to perform rack excitation on a rack system to measure a response signal, and the relation between the excitation signal and the response signal, namely, the frequency response function H (f) of the system is obtained through linear fitting according to the excitation signal and the measured response signal. The rack system consists of a rack and a test car.
The load spectrum data of the first hub motor angle module is a target driving signal and cannot be directly used for driving a rack to carry out a test, the load spectrum data of the first hub motor angle module is used as a response signal to be input into a rack system, and the driving signal to be input can be reversely calculated according to the following formula:
（1）
wherein X (f) is an input driving signal, and Y (f) is an output response signal.
Furthermore, as nonlinear relation exists in each link of the test system, and the frequency response function H (f) of the system is obtained by linear fitting, a larger error exists between the response signal and the target signal, which are calculated according to the linear system. In order to eliminate the effect of nonlinearity and reduce the error, it is necessary to correct the input drive signal by iterative processing so that the output response signal approaches the target signal. The iteration process comprises the following specific steps:
the iterative first driving spectrum is calculated, and the calculation formula is as follows:
（2）
（3）
wherein X is 0 (f) For driving the spectral frequency domain signal, α is the channel gain set for the first iteration, typically 1, x 0 (t) time domain signal of drive spectrum, IFFT [ X ] 0 (f)]Is an inverse fourier transform, used to convert the frequency domain signal into a time domain signal.
Using x 0 (t) performing rack excitation on the rack system as a current excitation signal to obtain a current response signal y of the system 0 (t) calculating the current response signal y 0 (t) and target Signal y T Error between (t):
（4）
（5）
wherein e 0 (t) is the time domain curve of the error signal, E 0 (f) The frequency domain signal, which is an error signal, is a fourier transform for converting the time domain signal into a frequency domain signal.
Obtaining a correction value of the current excitation signal according to the frequency domain signal of the error signal:
（6）
wherein,for the current excitation signal x 0 The correction value of (t), β being the gain during the iteration.
Obtaining next excitation signal x according to the current excitation signal and the correction value of the current excitation signal 1 (t)：
（7）
And continuing to perform rack excitation on the rack system by using the next excitation signal as a new excitation signal to obtain a new response signal of the system, and repeating the steps for iteration until the iteration index meets the iteration termination condition. And stopping iteration when the iteration index meets the iteration termination condition, and taking the current excitation signal as a target driving signal, namely a driving parameter for the complete vehicle fault test of the hub motor.
The calculation formula of the iteration index is as follows:
（8）
wherein E is RMS Is the root mean square value, D, of the difference between the target signal and the response signal RMS Is the root mean square error value of the target signal, and R is the average root mean square error value. Illustratively, as shown in Table 2, an iteration termination condition table is presented.
Table 2 an iteration termination condition table
As shown in table 2, the control channel refers to a data channel used in iterative calculation of the gantry; the average root mean square error refers to the convergence degree or the recurrence degree of iterative data compared with original data, and can also be regarded as an accuracy error; the relative damage ratio refers to the ratio of the damage caused by iterated data to the component and the damage caused by original data to the component, and the closer the relative damage ratio is to 1, the more accurate the relative damage ratio is. Illustratively, when the calculation channel is the main channel, R is less than or equal to 15% and meets the iteration termination condition, and when the calculation channel is the general channel, R is less than or equal to 25% and meets the iteration termination condition.
Furthermore, before the whole wheel hub motor fault test is carried out, the wheel hub motor needs to be ground in order to ensure that the wheel hub motor fully reaches the state of a heat engine and ensure that all components can work normally; running in may include forward and reverse rotation at a rated rotational speed and a rated torque. For example, the break-in duration may be 1h.
S2', developing a wheel hub motor fault test, and recording fault analysis data of a wheel hub motor angle module.
S21', carrying out a complete vehicle fault test of the hub motor according to the driving parameters, and collecting load spectrum data of a second hub motor angle module in the test process.
S22', extracting a second load index according to the load spectrum data of the second hub motor angle module.
And S23', after the test is finished, recording fault analysis data of the hub motor angle module.
S3', determining a weight matrix of the basic data and the fault analysis data according to the basic data and the fault analysis data.
S31', constructing a first weight matrix W according to the basic data of the vehicle and the first load index 1 。
Specifically, by changing the basic data of the vehicle, a first weight moment can be constructed by adopting a weight analysis method such as a CRITIC weight method, an entropy value method and the like and an optimization method thereofArray W 1 . Illustratively, consider the CRITIC weighting method as an example:
there are N pieces of vehicle basic data, M pieces of first load indexes, and M pieces of data sets related to the vehicle basic data for the 1 st load index, such as a six component peak of the wheel center. Illustratively, under the same test conditions, the basic parameters of the vehicle are changed for test, and m basic parameters of the vehicle are obtained through m combinations, and the following assumption is made: vehicle mass to test 5 loads, 4 tire models, 3 tire pressures, then m=5×4×3=60; the parameters of centroid position, corner module mass, corner module stiffness, etc. may also be varied to increase the number of data sets of the vehicle base data. And constructing an initial matrix X according to the vehicle basic data, the first load indexes and the M data sets related to the vehicle basic data, wherein M initial matrices are required to be constructed.
Taking the first load index as an example, the calculation formula of the initial matrix X is as follows:
（9）
wherein, the 1 st column to the nth column respectively represent the numerical value of the basic data of the vehicle, the kth column represents the numerical value of the first load index, k=n+1, and m represents the data line number.
Carrying out data standardization processing on the initial matrix X to obtain a standardized matrix Y, wherein for a first load index of a forward index, namely, an index with a better value of the obtained first load index, the larger value of the basic data of the vehicle, adopting a formula (10); for the first load index of the negative index, that is, the smaller the value of the vehicle base data is, the better the value of the obtained first load index is, the formula (11) is adopted:
（10）
（11）
wherein y is ij To normalize the data in matrix Y, x ij To initialize the data in matrix X, X j For initializing the j-th column of data in matrix X.
Calculating the average value of the j-th column in the initialization matrix XAnd standard deviation s j ，j=1,2，...，N，k：
（12）
（13）
According to the average valueAnd standard deviation s j Calculating the coefficient of variation v j ：
（14）
Calculating a correlation coefficient matrix R:
（15）
wherein r is jk Is a specific numerical value of a correlation coefficient matrix R, R jk For the association of the jth vehicle base data with the first load index k, j has a value in the range 1,2, N,is the covariance of the jth vehicle base data with the first load index k.
Then calculate the index information amount：
（16）
Finally, the weight W of the first load index is obtained j ：
（17）
Repeating the steps to obtain weights of M first load indexes, and constructing a first weight matrix W according to the weights of M first load indexes 1 。
S32', constructing a second weight matrix W according to the second load index and the fault analysis data 2 。
Specifically, a second weight matrix W may be constructed by using a weight analysis method such as CRITIC weight method, entropy method, and the like, and an optimization method thereof 2 . Second weight matrix W 2 Construction method of (a) and first weight matrix W 1 The construction method of (2) is consistent and will not be described in detail herein.
S33' according to the first weight matrix W 1 And a second weight matrix W 2 A third weight matrix W for deriving vehicle base data and fault analysis data 3 。
Specifically, a third weight matrix W 3 The calculation formula of (2) is as follows: w (W) 3 =W 1 ·W 2 According to a first weight matrix W 1 And a second weight matrix W 2 A third weight matrix W for deriving vehicle base data and fault analysis data 3 。
S4' taking the vehicle base data as input, and taking the third weight matrix W 3 And inputting the initial weight into a neural network prediction model to predict the fault condition of the hub motor to be tested.
Specifically, the structure of the neural network prediction model may include: setting the number of nodes of an input layer of the neural network prediction model as N, wherein N is the number of basic data of the vehicle, setting the number of nodes of an output layer of the neural network prediction model as P, and P is the number of fault analysis data; setting the number of hidden layers as N1, if the number of hidden layers is too large, the fitting phenomenon will occur, and the running speed and the accuracy of the model are affected, so that N1 can be set to be 1 or 2, the number of nodes of each layer is N2, and the number of N2 can be obtained by adopting a minimum training frequency method.
Further, an empirical value n21 of the node number can be obtained by using an empirical formula (18):
N2=（N+P） 1/2 +b（18）
wherein b is a natural number between [1,10 ].
Then an empirical value n22 of the node number is obtained according to the empirical formula (19):
N2=2N+1（19）
and obtaining a natural number of N2 between [ N21, N22], selecting a numerical value N23 of the node number by adopting a golden section method, calculating training times when the neural network model with the node number of N21, N22 and N23 is stable, selecting the settlement number corresponding to the minimum two times, continuing to take the value between corresponding nodes by adopting the golden section method, and repeating the steps until a unique numerical value is selected, namely the node number N2 of the hidden layer. By adopting the method to select proper node number, the result prediction can be rapidly realized, and the efficiency is improved.
Further, fig. 3 is a logic diagram of a prediction method of a neural network prediction model according to an embodiment of the present invention, referring to fig. 3, vehicle basic data is taken as input, and a third weight matrix W 3 As an initialization weight from the input layer to the hidden layer, the fault analysis data recorded after the test is used as training data. And establishing a three-layer or four-layer BP neural network model by utilizing MATLAB, training, and setting a training target as an error between a predicted value and an actual value, for example, 0.01. The tangent S-shaped tanig function is selected as an excitation function from an input layer to an hidden layer, the Purelin function is selected as an excitation function from the hidden layer to an output layer, the training frequency is set to be a larger value, and exemplarily, 10000 times can be set, the model can be converged without reaching the target training frequency, the learning rate is 0.01, and the stability of the system can be ensured by the smaller learning rate. After training, the network model successfully converges to the training target, at the moment, the BP neural network model training is completed, and the training times are recorded.
Further, the vehicle basic data of the vehicle model to be tested is taken as input, and a third weight matrix W is taken 3 As an initialized weight is input into a trained neural network prediction model, the fault condition of the hub motor to be tested can be obtained.
In some embodiments, a second load index may also be used as input, a second weight matrix W 2 And inputting the initialized weight into a trained neural network prediction model, and predicting the fault condition of the hub motor to be tested. The second load index and the basic data of the vehicle can be simultaneously used as input, and a second weight matrix W 2 And a third weight matrix W 3 And inputting the initialized weight into a trained neural network prediction model, and predicting the fault condition of the hub motor to be tested. By matching different initialization weights according to various input parameters, the compatibility and universality of the prediction model are improved.
According to the embodiment of the invention, the weight matrix of the vehicle basic data and the fault analysis data is determined by collecting the vehicle basic data required by the whole vehicle fault test of the hub motor and the fault analysis data of the hub motor angle module, the fault condition of the hub motor to be tested is obtained based on the neural network prediction model, the fault condition of the hub motor can be predicted only under the condition that the vehicle basic data is known, and the use requirement of the hub motor for long-term stable service under the whole vehicle running condition can be rapidly and effectively verified.
Fig. 4 is a flowchart of a fault test of an in-wheel motor corner module according to an embodiment of the present invention, referring to fig. 4, when the in-wheel motor fault test is an in-wheel motor corner module fault test, the method includes the following steps:
s1', acquiring angle module basic data required by a wheel hub motor fault test.
In particular, the corner module base data may include corner module mass, corner module centroid position, corner module stiffness, corner module deflection, suspension type, bar system length, material type, tire pressure, tire type, and the like.
Further, determining a standard working condition of a wheel hub motor angle module fault test; the standard conditions include a start condition, a drive condition, a speed change condition, a steering condition, and a braking condition. The chassis dynamometer program is developed according to the standard working conditions, and the sequence of the tests of the standard working conditions can be randomly ordered according to requirements.
S2'' carrying out a wheel hub motor angle module fault test according to the standard working condition, and recording fault analysis data of the wheel hub motor angle module.
Specifically, the fault analysis data of the wheel hub motor angle module include whether the starting function, the driving performance, the steering and braking functions of the angle module are good, whether the heat dissipation and the sealing performance are good, whether the motor control system and the control logic are normal, whether the basic function is lost, the performance is greatly declined and the like, and simultaneously, the time of occurrence of faults, the corresponding driving mileage and the like are recorded.
The requirements of each of the operating conditions are illustratively described with respect to the test sequence of the start-up operating condition test, the drive operating condition test, the shift operating condition test, the steering operating condition test, and the brake operating condition test.
1) And (3) starting working condition test: according to the starting step of the hub motor, a starting working condition test of the hub motor angle module is carried out, and because the time for stabilizing the motor starting is about 5s, the motor is maintained for 5s after the hub motor is successfully started, and is continuously started for 10 times as one cycle, 200 times are required for starting, namely 20 cycles, and the motor is stopped for 5min in the middle, so that abnormal faults caused by the continuous starting of the motor are prevented, and statistical errors occur.
2) Driving condition test: FIG. 5 is a schematic diagram of a driving condition of an in-wheel motor angle module according to an embodiment of the present invention, referring to FIG. 5, three driving speeds n are set 1 、n 2 And n max Motor speeds corresponding to 60km/h, 80km/h and highest cruise speed, respectively. Each cycle 14h, comprising: 60km/h for 4h,80km/h for 4h, highest cruising speed for 4h, and stopping slowly for 2h at a constant speed in the middle. A total of 5 cycles, 70h, were measured.
3) Speed change working condition test: fig. 6 is a schematic diagram of a variable speed working condition of an angle module of a hub motor according to an embodiment of the present invention, referring to fig. 6, the variable speed working condition is composed of 11 cycles, each cycle has a test mileage of 6km, and a total test mileage of 5000km. The first 9 loops are composed of a starting section, a decelerating section, an idling section and a deceleration-to-stop section. The starting section comprises starting acceleration to the circulation maximum vehicle speed and driving at a constant speed of 500m; the deceleration section comprises the steps of decelerating to a preset speed, such as 32km/h, accelerating to a circulating maximum speed, and driving at a constant speed of 800m; the idle section includes idle 15s and then accelerates to a cyclical maximum vehicle speed, traveling at a constant speed of 500m. For example, the driving conditions of the first 9 cycles are all from a start section, a deceleration section, an idle section, a deceleration section, and a deceleration to a stop, and the total mileage of each cycle is 6km. The 10 th cycle may be operated at a constant speed of 89km/h for a constant speed of 6km and then decelerated to a stop. The 11 th cycle can be started firstly, accelerated to the highest speed by the maximum acceleration, run at a constant speed for 3km, then decelerated and stopped, idle for 15s, started again, accelerated to the highest speed by the maximum acceleration, run at a constant speed for 3km, then decelerated to a stop, and the total mileage is 6km.
4) Steering condition test: the steering working condition test is divided into two parts, namely an 8-shaped steering test and an in-situ steering test (comprising 90-degree steering), which are sequentially carried out. The 8-word steering test track refers to the lemniscate path in the steering portability test in the GB/T6323-2014 standard, and is stopped for 30s midway for 200 times, namely 20 times in each cycle. The in-situ steering test is performed according to the sequential operations of "wheel alignment, wheel left limit (or 90 degrees), wheel alignment, wheel right limit (or 90 degrees), wheel alignment, each cycle was stopped for 30s in the middle of 10 times, and 200 times, namely 20 times.
5) Brake condition test: the braking condition test is divided into two parts, namely an emergency braking test and a continuous acceleration and deceleration test, which are carried out in sequence. The emergency braking test is made forward in an emergency mode at the speed of 60km/h, and is braked backward in an emergency mode at the speed of 15km/h, each cycle is stopped for 10 times, the midway is stopped for 15min, and the total number of times is 300, namely 30 cycles; wherein, 60km/h and 15km/h are industriesThe emergency braking test vehicle speed is commonly used. FIG. 7 is a schematic diagram of a continuous acceleration/deceleration condition of an angular module of a hub motor according to an embodiment of the present invention, referring to FIG. 7, the continuous acceleration/deceleration test is to uniformly accelerate to n respectively 1 、n 2 And n max Then uniformly decelerating to 0; wherein n is 1 、n 2 And n max Motor speeds corresponding to 60km/h, 80km/h and highest cruise speed, respectively. Each cycle is stopped for 15min in the middle of 10 times, and 100 times are 10 times.
S3'' constructing a fourth weight matrix W according to the corner module basic data and the fault analysis data 4 。
Specifically, a fourth weight matrix W may be constructed by using a weight analysis method such as CRITIC weight method, entropy method, and the like, and an optimization method thereof 4 . Fourth weight matrix W 4 Construction method of (a) and first weight matrix W 1 The construction method of (2) is consistent and will not be described in detail herein.
S4'', taking the corner module basic data as input and taking a fourth weight matrix W 4 And inputting the initial weight into a neural network prediction model to predict the fault condition of the hub motor to be tested.
Specifically, the neural network prediction model of the wheel hub motor angle module fault test is the same as the neural network prediction model of the wheel hub motor whole vehicle fault test in principle, and will not be described in detail herein.
According to the embodiment of the invention, the failure analysis data of the angle module of the hub motor is collected, the weight matrix of the basic data and the failure analysis data is determined, the failure condition of the hub motor to be tested is obtained based on the neural network prediction model, the failure condition of the hub motor can be predicted only under the condition that only the basic data of the angle module is known, and the use requirement of the hub motor for long-term stable service under the running conditions of the whole vehicle and the independent angle module can be rapidly and effectively verified.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of the present application. As used in this specification, the terms "a," "an," "the," and/or "the" are not intended to be limiting, but rather are to be construed as covering the singular and the plural, unless the context clearly dictates otherwise. The terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method or apparatus comprising such elements.
It should also be noted that the positional or positional relationship indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. are based on the positional or positional relationship shown in the drawings, are merely for convenience of describing the present invention and simplifying the description, and do not indicate or imply that the apparatus or element in question must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention. Unless specifically stated or limited otherwise, the terms "mounted," "connected," and the like are to be construed broadly and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention will be understood in specific cases by those of ordinary skill in the art.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some or all of the technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.
Claims (3)
1. The method for predicting the failure condition of the hub motor is characterized by comprising the following steps of:
s1, acquiring basic data required by a wheel hub motor fault test; the wheel hub motor fault test comprises a wheel hub motor whole vehicle fault test or a wheel hub motor angle module fault test; when the wheel hub motor fault test is a wheel hub motor complete vehicle fault test, the basic data comprise vehicle basic data, and when the wheel hub motor fault test is a wheel hub motor angle module fault test, the basic data comprise angle module basic data;
s2, developing a wheel hub motor fault test, and recording fault analysis data of a wheel hub motor angle module;
s3, determining a weight matrix of the basic data and the fault analysis data according to the basic data and the fault analysis data;
s4, inputting basic data of the hub motor to be tested and the weight matrix into a neural network prediction model, and predicting the fault condition of the hub motor to be tested;
when the in-wheel motor fault test is an in-wheel motor complete vehicle fault test, after the step S1, the method further includes:
collecting load spectrum data of a first hub motor angle module required by a whole-vehicle fault test of the hub motor, and extracting a first load index according to the load spectrum data of the first hub motor angle module;
calculating driving parameters for the complete vehicle fault test of the hub motor according to the load spectrum data of the first hub motor angle module;
s2, carrying out a wheel hub motor fault test, recording fault analysis data of a wheel hub motor angle module, wherein the fault analysis data comprise:
s21', carrying out a complete vehicle fault test of the hub motor according to the driving parameters, and collecting load spectrum data of a second hub motor angle module in the test process;
s22', extracting a second load index according to the load spectrum data of the second hub motor angle module;
s23', after the test is finished, recording fault analysis data of the hub motor angle module;
the step S3 of determining the weight matrix of the basic data and the fault analysis data according to the basic data and the fault analysis data comprises the following steps:
s31', constructing a first weight matrix W according to the vehicle basic data and the first load index 1 ；
S32', constructing a second weight matrix W according to the second load index and the fault analysis data 2 ；
S33' according to the first weight matrix W 1 And the second weight matrix W 2 A third weight matrix W for deriving vehicle base data and fault analysis data 3 ；
When the in-wheel motor fault test is an in-wheel motor angle module fault test, after the step S1, the method further includes:
determining a standard working condition of the hub motor angle module fault test; the standard working conditions comprise a starting working condition, a driving working condition, a speed changing working condition, a steering working condition and a braking working condition;
s2, carrying out a wheel hub motor fault test, recording fault analysis data of a wheel hub motor angle module, wherein the fault analysis data comprise:
developing a wheel hub motor angle module fault test according to the standard working condition, and recording fault analysis data of the wheel hub motor angle module;
the step S3 of determining the weight matrix of the basic data and the fault analysis data according to the basic data and the fault analysis data comprises the following steps:
constructing a fourth weight matrix W according to the angle module basic data and the fault analysis data 4 。
2. The method for predicting the failure condition of an in-wheel motor according to claim 1, wherein when the in-wheel motor failure test is an in-wheel motor complete vehicle failure test, the step S4 of inputting the basic data of the in-wheel motor to be detected and the weight matrix into a neural network prediction model, the predicting the failure condition of the in-wheel motor to be detected includes:
taking the vehicle basic data as input, taking the third weight matrix W 3 And inputting the initial weight into the neural network prediction model to predict the fault condition of the hub motor to be detected.
3. The method for predicting the failure condition of an in-wheel motor according to claim 1, wherein when the in-wheel motor failure test is an in-wheel motor angle module failure test, the step S4 of inputting the basic data of the in-wheel motor to be detected and the weight matrix into a neural network prediction model, the predicting the failure condition of the in-wheel motor to be detected includes:
taking the corner module basic data as input, taking the fourth weight matrix W 4 And inputting the initial weight into a neural network prediction model to predict the fault condition of the hub motor to be detected.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311656884.1A CN117390519B (en) | 2023-12-06 | 2023-12-06 | Wheel hub motor fault condition prediction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311656884.1A CN117390519B (en) | 2023-12-06 | 2023-12-06 | Wheel hub motor fault condition prediction method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117390519A CN117390519A (en) | 2024-01-12 |
| CN117390519B true CN117390519B (en) | 2024-04-09 |
Family
ID=89472409
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311656884.1A Active CN117390519B (en) | 2023-12-06 | 2023-12-06 | Wheel hub motor fault condition prediction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117390519B (en) |
Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106769093A (en) * | 2016-12-22 | 2017-05-31 | 北京汽车股份有限公司 | Road based on wheel six square phase load is made an uproar prediction analysis method and system |
| CN108918137A (en) * | 2018-06-08 | 2018-11-30 | 华北水利水电大学 | Fault Diagnosis of Gear Case devices and methods therefor based on improved WPA-BP neural network |
| CN110031225A (en) * | 2019-03-15 | 2019-07-19 | 江苏大学 | A kind of hub motor bearing failure based on artificial hydrocarbon network gradually diagnostic method |
| CN110865628A (en) * | 2019-10-25 | 2020-03-06 | 清华大学深圳国际研究生院 | New energy automobile electric control system fault prediction method based on working condition data |
| CN111859290A (en) * | 2020-06-18 | 2020-10-30 | 江苏大学 | Wheel hub motor mechanical fault diagnosis method based on rough set and artificial hydrocarbon network |
| CN111881717A (en) * | 2020-06-08 | 2020-11-03 | 江苏大学 | Hub motor state identification method based on Bayesian network and improved DS theory |
| CN112373293A (en) * | 2020-09-11 | 2021-02-19 | 东风越野车有限公司 | Fault processing method for distributed driving system of hub motor |
| CN112611584A (en) * | 2020-05-18 | 2021-04-06 | 北京金风慧能技术有限公司 | Fatigue failure detection method, device, equipment and medium for wind generating set |
| CN113468728A (en) * | 2021-06-11 | 2021-10-01 | 国电南京自动化股份有限公司 | Variable pitch system fault prediction method based on neural network |
| WO2021218277A1 (en) * | 2020-04-27 | 2021-11-04 | 湖南大学 | Analysis and test method for fault diagnosis of vehicle control unit of electric vehicle |
| WO2021223594A1 (en) * | 2020-05-07 | 2021-11-11 | 北京理工大学 | Multi-objective optimization method and system for driving system of distributed driving electric vehicle |
| CN113919102A (en) * | 2021-11-08 | 2022-01-11 | 浙江理工大学 | Method and system for analyzing load transfer characteristics of main transmission system of wind turbine generator |
| WO2023283667A1 (en) * | 2021-07-14 | 2023-01-19 | Stop-In-Time Gmbh | Method for braking a vehicle |
| CN115903482A (en) * | 2022-11-07 | 2023-04-04 | 华南理工大学 | Adaptive neural network fault-tolerant cooperative control method of multi-flexible-arm system |
| CN116148104A (en) * | 2023-04-18 | 2023-05-23 | 中汽研汽车检验中心（天津）有限公司 | Method and equipment for determining load spectrum of wheel double-shaft fatigue test based on actual working condition |
| CN116561514A (en) * | 2023-05-11 | 2023-08-08 | 太仓市凯福士机械有限公司 | Method, system, device and medium for diagnosing faults of vehicle hub bearing unit |
| CN116720104A (en) * | 2023-05-30 | 2023-09-08 | 重庆工商大学 | Wind driven generator gear box output shaft fault prediction method |
| CN116771610A (en) * | 2023-05-19 | 2023-09-19 | 华北电力大学 | Method for adjusting fault evaluation value of variable pitch system of wind turbine |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20210061027A1 (en) * | 2017-10-26 | 2021-03-04 | Applied Mechatronic Products | Apparatus and method for vehicular monitoring, analysis, and control |
| CN111845734B (en) * | 2020-07-31 | 2021-03-02 | 北京理工大学 | Fault-tolerant tracking control method for four-wheel distributed electrically-driven automatic driving vehicle |
| CN114614720B (en) * | 2022-03-01 | 2022-09-20 | 浙江科技学院 | Fault-tolerant decision-making method and system for vehicle hub driving system sensor failure |
-
2023
- 2023-12-06 CN CN202311656884.1A patent/CN117390519B/en active Active
Patent Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106769093A (en) * | 2016-12-22 | 2017-05-31 | 北京汽车股份有限公司 | Road based on wheel six square phase load is made an uproar prediction analysis method and system |
| CN108918137A (en) * | 2018-06-08 | 2018-11-30 | 华北水利水电大学 | Fault Diagnosis of Gear Case devices and methods therefor based on improved WPA-BP neural network |
| CN110031225A (en) * | 2019-03-15 | 2019-07-19 | 江苏大学 | A kind of hub motor bearing failure based on artificial hydrocarbon network gradually diagnostic method |
| CN110865628A (en) * | 2019-10-25 | 2020-03-06 | 清华大学深圳国际研究生院 | New energy automobile electric control system fault prediction method based on working condition data |
| WO2021218277A1 (en) * | 2020-04-27 | 2021-11-04 | 湖南大学 | Analysis and test method for fault diagnosis of vehicle control unit of electric vehicle |
| WO2021223594A1 (en) * | 2020-05-07 | 2021-11-11 | 北京理工大学 | Multi-objective optimization method and system for driving system of distributed driving electric vehicle |
| CN112611584A (en) * | 2020-05-18 | 2021-04-06 | 北京金风慧能技术有限公司 | Fatigue failure detection method, device, equipment and medium for wind generating set |
| CN111881717A (en) * | 2020-06-08 | 2020-11-03 | 江苏大学 | Hub motor state identification method based on Bayesian network and improved DS theory |
| CN111859290A (en) * | 2020-06-18 | 2020-10-30 | 江苏大学 | Wheel hub motor mechanical fault diagnosis method based on rough set and artificial hydrocarbon network |
| CN112373293A (en) * | 2020-09-11 | 2021-02-19 | 东风越野车有限公司 | Fault processing method for distributed driving system of hub motor |
| CN113468728A (en) * | 2021-06-11 | 2021-10-01 | 国电南京自动化股份有限公司 | Variable pitch system fault prediction method based on neural network |
| WO2023283667A1 (en) * | 2021-07-14 | 2023-01-19 | Stop-In-Time Gmbh | Method for braking a vehicle |
| CN113919102A (en) * | 2021-11-08 | 2022-01-11 | 浙江理工大学 | Method and system for analyzing load transfer characteristics of main transmission system of wind turbine generator |
| CN115903482A (en) * | 2022-11-07 | 2023-04-04 | 华南理工大学 | Adaptive neural network fault-tolerant cooperative control method of multi-flexible-arm system |
| CN116148104A (en) * | 2023-04-18 | 2023-05-23 | 中汽研汽车检验中心（天津）有限公司 | Method and equipment for determining load spectrum of wheel double-shaft fatigue test based on actual working condition |
| CN116561514A (en) * | 2023-05-11 | 2023-08-08 | 太仓市凯福士机械有限公司 | Method, system, device and medium for diagnosing faults of vehicle hub bearing unit |
| CN116771610A (en) * | 2023-05-19 | 2023-09-19 | 华北电力大学 | Method for adjusting fault evaluation value of variable pitch system of wind turbine |
| CN116720104A (en) * | 2023-05-30 | 2023-09-08 | 重庆工商大学 | Wind driven generator gear box output shaft fault prediction method |
Non-Patent Citations (6)
| Title |
|---|
| State Feedback Control for a PM Hub Motor Based on Gray Wolf Optimization Algorithm;Xiaodong Sun 等;《IEEE TRANSACTIONS ON POWER ELECTRONICS》;第35卷(第1期);全文 * |
| 基于DBNs的轮毂电机机械故障在线诊断方法;李仲兴;陈震宇;薛红涛;殷苏群;江洪;;振动.测试与诊断(04);全文 * |
| 基于数据驱动的轮毂电机轴承故障诊断方法研究;周宇;《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》(第02期);全文 * |
| 基于深度学习的直驱风电机组故障辨识;丁显;韩宁宁;滕伟;;可再生能源(10);全文 * |
| 基于稀疏表示的轮毂电机轴承故障诊断方法研究;周庄;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》(第11期);全文 * |
| 设备故障趋势预测的分析与应用;续媛君;潘宏侠;;振动、测试与诊断;20060130(04);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117390519A (en) | 2024-01-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109357822B (en) | Bridge rapid testing and evaluating method based on time-varying power characteristic change of axle coupling system | |
| US6766230B1 (en) | Model-based fault detection and isolation system and method | |
| CN113092115B (en) | Digital twin model construction method of digital-analog combined drive full-life rolling bearing | |
| JP2009536736A (en) | Vehicle testing and simulation using embedded simulation models and physical components | |
| US6223120B1 (en) | Cylinder torque estimation using crankshaft angular response measurements | |
| CN109813511B (en) | Bridge rapid test and parameter identification method based on mobile vehicle | |
| CN102944422B (en) | Driving load fatigue representing and testing method for transmission system of automobile | |
| US6687585B1 (en) | Fault detection and isolation system and method | |
| CN114115200B (en) | NVH (noise and harshness) testing system and testing method for electric drive assembly of electric automobile | |
| US20060282197A1 (en) | Method for optimizing vehicles and engines used for driving such vehicles | |
| CN117390519B (en) | Wheel hub motor fault condition prediction method | |
| CN115270296A (en) | Method and system for analyzing fatigue durability of commercial vehicle cab | |
| CN113515808B (en) | Noise prediction method of power assembly | |
| Liu et al. | Reliability analysis and evaluation of a brake system based on competing risk | |
| CN112560162B (en) | Method and device for reducing suspended road spectrum load of power assembly | |
| CN115481488B (en) | Fang Zhui type energy-absorbing structure crashworthiness multi-objective optimization method based on machine learning | |
| CN114755027B (en) | Whole vehicle multiaxial loading test bed, test method and medium | |
| CN108437798B (en) | Electric scooter fault diagnosis and estimation method based on bonding diagram model | |
| CN114861434A (en) | Method for checking precision of vehicle multi-body dynamic model and storage medium | |
| CN112733887A (en) | Method for detecting fault of hub motor of electric vehicle driven by supervision data | |
| CN111241724B (en) | Fatigue life prediction method for wide-body mining vehicle frame | |
| CN113155486A (en) | Durability simulation test method and system for power assembly suspension system | |
| Lv et al. | State estimation of cyber-physical vehicle systems | |
| Pisu et al. | Model Based Diagnostics for Vehicle Systems | |
| CN116400674A (en) | Whole-vehicle fault modeling and diagnosing method and system for hybrid electric vehicle |
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 |