CN113821866A - Method for predicting residual life of finger protection adhesive tape of urban rail door system - Google Patents
Method for predicting residual life of finger protection adhesive tape of urban rail door system Download PDFInfo
- Publication number
- CN113821866A CN113821866A CN202010563092.XA CN202010563092A CN113821866A CN 113821866 A CN113821866 A CN 113821866A CN 202010563092 A CN202010563092 A CN 202010563092A CN 113821866 A CN113821866 A CN 113821866A
- Authority
- CN
- China
- Prior art keywords
- adhesive tape
- finger protection
- degradation state
- state
- predicting
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 239000002390 adhesive tape Substances 0.000 title claims abstract description 60
- 238000000034 method Methods 0.000 title claims abstract description 31
- 238000006731 degradation reaction Methods 0.000 claims abstract description 53
- 230000015556 catabolic process Effects 0.000 claims abstract description 51
- 238000013145 classification model Methods 0.000 claims abstract description 20
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 9
- 229920001971 elastomer Polymers 0.000 claims description 47
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000012360 testing method Methods 0.000 claims description 8
- 238000007635 classification algorithm Methods 0.000 claims description 7
- 238000012549 training Methods 0.000 claims description 7
- 238000000354 decomposition reaction Methods 0.000 claims description 6
- 230000035945 sensitivity Effects 0.000 claims description 6
- 238000010219 correlation analysis Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 4
- 238000001125 extrusion Methods 0.000 claims description 3
- 150000001875 compounds Chemical class 0.000 claims description 2
- 238000010206 sensitivity analysis Methods 0.000 claims description 2
- 238000012423 maintenance Methods 0.000 abstract description 6
- 238000001514 detection method Methods 0.000 abstract description 3
- 230000001066 destructive effect Effects 0.000 abstract description 2
- 238000012544 monitoring process Methods 0.000 abstract description 2
- 230000002349 favourable effect Effects 0.000 abstract 1
- 230000000875 corresponding effect Effects 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 210000002569 neuron Anatomy 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 3
- 229920002943 EPDM rubber Polymers 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000007789 sealing Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000008358 core component Substances 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000013075 data extraction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/15—Vehicle, aircraft or watercraft design
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/04—Ageing analysis or optimisation against ageing
Abstract
The invention discloses a method for predicting the residual life of an urban rail door system finger protection adhesive tape, which comprises the steps of extracting the characteristics of motor current curves corresponding to the adhesive tape in different operation time periods, establishing an adhesive tape degradation state classification model based on an intelligent algorithm, identifying the degradation state of the current finger protection adhesive tape by using the classification model, predicting the residual life of the adhesive tape and judging whether the adhesive tape needs to be replaced. The method does not need to collect finger protection adhesive tape samples for destructive detection, can realize real-time monitoring of the state of the adhesive tape, is favorable for deeply knowing the degradation process of the adhesive tape, scientifically arranges the replacement period and reduces the maintenance cost.
Description
Technical Field
The invention relates to a method for predicting the residual life of a finger protection rubber strip, in particular to a method for predicting the residual life of a finger protection rubber strip of an urban rail door system.
Background
With the rapid development of the rail transit industry, the safety and reliability of the urban rail vehicle door are generally concerned by people. The finger protection rubber strip is one of the core components of the door system of the railway vehicle, and when the door is closed in place, the rubber strips are mutually extruded to realize the sealing effect of the door system. The finger protection rubber strip is made of Ethylene Propylene Diene Monomer (EPDM), and aging failure can occur under the influence of environment in the using process, so that the sealing effect of the car door is reduced. At present, the urban rail vehicle door finger protection adhesive tape is maintained in a regular replacement mode, the replacement frequency is high, the whole life cycle management of the adhesive tape cannot be realized, and the operation and maintenance cost is high. The service life of the finger protection rubber strip is predicted to obtain the degradation state of the rubber strip, and the reasonable maintenance period arrangement is facilitated.
The method adopted for predicting the service life of the finger protection rubber strip at present is to take old finger protection rubber strips with different service times as samples, detect mechanical performance indexes such as hardness, tensile strength, elongation at break and the like of the rubber strips, calculate the performance retention rate of the rubber strips, perform linear or nonlinear fitting on the detected mechanical performance data, and calculate the service life of the finger protection rubber strip according to a fitting equation. The method needs to collect finger protection rubber strip samples at different stages, and usually needs a longer test period and higher test cost.
The traditional life prediction method is based on sample mechanical property data fitting life curve, the finger protection rubber strips under different service time are collected for testing, sample collection is random, the rubber strip state cannot be continuously monitored, and different interpolation methods have different curve fitting effects, so that the life prediction is not accurate enough.
Disclosure of Invention
The purpose of the invention is as follows: the invention aims to provide a method for predicting the residual life of finger protection rubber strips of an urban rail door system, which is beneficial to scientifically arranging a replacement period and reducing maintenance cost.
The technical scheme is as follows: the method for predicting the residual life of the finger protection rubber strip of the urban rail door system comprises the following steps of:
(1) collecting door closing current data of the adhesive tape in the whole life cycle operation, and intercepting a motor current curve segment during the extrusion work of the adhesive tape;
(2) extracting the characteristic values of a time domain and a frequency domain of the current curve to form a characteristic vector;
(3) performing correlation analysis and sensitivity analysis on the extracted features to remove redundant features;
(4) calculating Fisher criterion scores for the time domain and frequency domain characteristic indexes, and selecting characteristics with higher scores;
(5) dividing the whole life cycle of the rubber strip into N types of degradation states according to time: learning time domain and frequency domain characteristics of a current curve of the adhesive tape in a full life cycle by using a classification algorithm under a normal state, a degradation state 1, a degradation state 2, a degradation state 3 … …, a degradation state N-2 and a failure state to obtain an adhesive tape degradation state classification model, wherein the normal state, the degradation state 1, the degradation state 2, the degradation state 3 … …, the degradation state N-2 and the failure state are determined according to the service time, and N is more than 4 and less than 10;
(6) and identifying the degradation state of the current finger-protecting adhesive tape according to the adhesive tape degradation state classification model, calculating the residual life of the adhesive tape, and judging whether the adhesive tape needs to be replaced.
Further, in step (2), the time-domain features include a maximum, a minimum, a mean, a variance, a skewness, a kurtosis, and a kurtosis. The extraction method of the frequency domain features comprises the steps of carrying out wavelet decomposition on motor current data to obtain energy of each sub-frequency band and determining a frequency domain feature set.
In the step (3), the feature correlation analysis calculation formula is as follows:
in the formula, ρx,yIs the characteristic correlation, x and y are characteristic values,
the standard deviation is corresponding to the characteristic value.
In the step (3), the characteristic sensitivity calculation formula is as follows:
in the formula, λ (x)j) Is xjF (x) is the output value of the classification model, ST is the learning sample, xjIs the characteristic taken.
In the step (4), the Fisher criterion score calculation formula is as follows:
in the formula (I), the compound is shown in the specification,
is a Fisher-Tropsch discrimination score and,
are respectively a characteristic f1At the average value of the sample space P, Q,
is the corresponding variance.
In the step (5), an intelligent algorithm is used for constructing the adhesive tape degradation state classification model, and the method comprises the following steps:
and (4) outputting the N-2 degradation state labels as samples, forming a feature vector with the features selected in the step (4), randomly dividing the feature vector into learning samples and testing samples, inputting the learning samples into an intelligent algorithm for training to obtain an adhesive tape degradation state classification model, verifying the classification accuracy of the classification model by using the testing samples, and if the classification accuracy does not meet the set requirement, adjusting algorithm parameters until the classification accuracy meets the requirement.
In the step (6), the remaining life calculation formula is as follows:
wherein i is 1,2iFor the residual life of the adhesive tape in different states, T (k) is the residual life of the adhesive tape in the kth state, and k is whenFront strip status category.
In the door closing process of the door system, the finger protection rubber strip has the largest influence on the torque of the screw rod, so that the effect of the finger protection rubber strip can be indirectly reflected according to the change curve of the motor current, the current curve under the action of the finger protection rubber strip in different stages is subjected to feature extraction, a classification algorithm is utilized for learning, a rubber strip state degradation model is established, the current finger protection rubber strip state is classified based on the algorithm model, the current rubber strip degradation state is obtained, and whether the rubber strip needs to be replaced or not is judged.
Has the advantages that: compared with the prior art, the invention has the following remarkable advantages:
(1) the method has the advantages that destructive detection is performed without collecting finger protection adhesive tape samples, the state of the adhesive tape can be monitored in real time, the degradation process of the adhesive tape can be deeply known, the replacement period is reasonably arranged, replacement as required is realized, the maintenance cost is reduced, and the operation reliability is improved;
(2) and (4) researching the degradation state of the finger protection rubber strip in different stages based on the motor current characteristics. Dividing the whole life cycle of the rubber strip into 4 stages, extracting the current characteristics of each stage, establishing a finger protection rubber strip degradation state classification model by using a classification algorithm, classifying and identifying the current finger protection rubber strip state based on the classification model, and judging whether the rubber strip needs to be replaced;
(3) based on the influence of the finger protection adhesive tape on the door closing current, the adhesive tape sample collection and detection test are not needed, the degradation state of the adhesive tape is researched by analyzing the current curve characteristic of the motor, the effective monitoring on the whole life cycle of the adhesive tape can be realized, and the finger protection adhesive tape life prediction method has important guiding significance for improving the accuracy of the finger protection adhesive tape life prediction and reasonably arranging the adhesive tape maintenance cycle.
Drawings
FIG. 1 is a flow chart of a method for predicting the residual life of a finger protection rubber strip of an urban rail door system;
FIG. 2 is a schematic view of the finger-protecting rubber strip installation position;
FIG. 3 is a section of a current curve corresponding to the finger guard strip;
fig. 4 is a BP neural network classification algorithm structure according to an embodiment of the present invention.
Detailed Description
The technical solution of the present invention is further illustrated by the following examples.
As shown in FIG. 1, the method for predicting the residual life of the finger protection rubber strip of the urban rail door system comprises the following specific steps:
(1) collecting door closing current data of the adhesive tape in the whole life cycle operation, and judging and intercepting a current curve segment when the adhesive tape is extruded.
As shown in fig. 2, the installation position of the finger protection rubber strip is schematically shown, in the door closing in place process, the left finger protection rubber strip and the right finger protection rubber strip are mutually extruded, the current of the motor is increased, and the current curve segment of the rubber strip in the extrusion stage is intercepted, as shown in fig. 3.
(2) And extracting characteristic values of the current curve.
Extracting time domain characteristics of the current curve, which mainly comprises the following steps: maximum, minimum, mean, variance, skewness, kurtosis, and the like.
Extracting the maximum value and the minimum value of the intercepted current curve segment to determine the range of the data change interval, wherein the expression is as follows:
xmax＝max|xi|
xmin＝min|xi|
i＝1,2,...,N
the mean expression for data extraction is as follows:
the variance of the current data reflects the degree of deviation of each data from the mean, and the specific expression is as follows:
skewness reflects the degree of asymmetry of the data, and the expression is as follows:
the kurtosis reflects the smoothness of the waveform, and the expression is as follows:
wherein XrmsIs the root-mean-square value of the signal,
kurtosis reflects the extreme degree of the peak in the waveform, and is expressed as follows:
in the formula, XpIs a peak.
The frequency domain feature extraction is mainly to perform wavelet decomposition on the motor current data to obtain the energy of each sub-band and determine a frequency domain feature set.
Based on a scale function
And a wavelet function ψ (t) defining two functions of the wavelet transform:
ω1(t)＝ψ(t)
the decomposition of the signal x (t) in one of the wavelet sub-vector spaces into signals
Wherein the content of the first and second substances,
wavelet packet coefficients corresponding to nodes (j, n)
The signal x (t) is wavelet decomposed into the following form:
ωn,j,k(t) is an orthogonal wavelet basis
Sub-band signal
The energy of (d) is calculated by:
and 3-layer wavelet decomposition is selected to perform wavelet decomposition on the current signal to obtain subband energy as the frequency domain energy characteristic of the adhesive tape degradation state.
(3) And (5) analyzing feature redundancy.
The selected feature value may have a feature value redundancy phenomenon. Although the redundant features also contain sample information and do not influence the correctness of the classification result, the removal of the redundant features is helpful for reducing the dimension of the feature value and improving the operation efficiency. The correlation of each feature is analyzed to compare how closely the variable factors are. Calculating a correlation coefficient between any two features on a learning sample, if the correlation coefficient is greater than a certain threshold value, indicating that the two features are strongly correlated and eliminating the features with small signal-to-noise ratio, wherein the correlation coefficient calculation formula is as follows:
the signal-to-noise ratio can be used to weigh the amount of classification information for a feature, i.e., the
Where d (m) is the signal-to-noise ratio, μ is the sample mean, and σ is the sample standard deviation. If the signal-to-noise ratio of the two samples is equal to 0, the feature is removed, but if the variance difference is large, the feature can still be used as a distinguishing feature, in order to avoid the phenomenon of the false removal, the signal-to-noise ratio formula is corrected, and the correction result is as follows:
d(m)＝|μ1-μ2|+ln|σ1-σ2|
further screening the residual characteristics through sensitivity, wherein the sensitivity calculation formula is as follows:
in the formula, λ (x)j) Is xjF (x) is the output value of the classification model, in this example the output of the BP neural network, and ST is the learning sample.
And calculating the sensitivity of the characteristic value according to the formula, and removing the characteristic value with the minimum sensitivity as a redundant characteristic.
(4) And (5) evaluating the characteristics.
Evaluating the obtained characteristics, wherein the evaluation method adopts Fisher criterion:
selecting the data mark of the normal state of the adhesive tape as Good, the data mark of the failure state of the adhesive tape as Bad, calculating the fisher discrimination score of each feature, and selecting the 6 types of feature values with the highest scores.
(5) And establishing an adhesive tape degradation state classification model.
In this embodiment, divide into the adhesive tape life cycle: the method comprises a normal state, a degradation state 1, a degradation state 2, a degradation state 3, a degradation state 4 and a failure state, wherein each stage corresponds to different labels, and a classification algorithm is used for learning the current curve characteristics of the adhesive tape in the whole life cycle to obtain an adhesive tape degradation state classification model.
In the present embodiment, the classification algorithm uses a BP neural network for classification training, and the algorithm structure is shown in fig. 4. The sigmoid function is adopted in the middle of the neuron for excitation, normalization processing needs to be carried out on input characteristic data before the function is used, the network convergence speed can be increased through the normalization processing, and the training efficiency is improved.
x' is normalized sample data, x is input sample data, max (x) and min (x) are the maximum and minimum values of the input features, respectively.
Setting the number of the implied neurons to be 4, the learning rate to be 0.001 and the expected error to be 10-6The maximum number of iterations is 1000.
The specific training process of the BP neural network model comprises the following steps: after the current curve characteristics are input, neurons of an input layer are activated firstly, then information is transmitted to each neuron of a hidden layer for information processing and transformation, and finally the transformed information is transmitted to an output layer for further processing and output; if the difference between the output result and the preset output value is larger than the given error, the neural network carries out reverse training and repeatedly corrects the threshold value of each layer of weight until the error value between the network output and the preset value reaches the target required error range requirement or the learning frequency is larger than the set maximum learning frequency, and the training is stopped.
(6) Diagnosing and identifying the degradation state of the current finger-protecting adhesive tape according to the adhesive tape degradation state classification model established in the S5, and judging the adhesive tape state, wherein the corresponding residual lives of the normal state, the degradation state 1, the degradation state 2, the degradation state 3, the degradation state 4 and the failure state are respectively T1、T2、T3、T4、T5、T6The residual life of the rubber strip in the current state is calculated by the following formula.
Claims (8)
1. A method for predicting the residual life of finger protection rubber strips of an urban rail door system is characterized by comprising the following steps:
(1) collecting door closing current data of the adhesive tape in the whole life cycle operation, and intercepting a motor current curve segment during the extrusion work of the adhesive tape;
(2) extracting the characteristic values of the time domain and the frequency domain of the current curve to form a characteristic vector;
(3) performing correlation analysis and sensitivity analysis on the extracted features to remove redundant features;
(4) calculating Fisher criterion scores for the time domain and frequency domain characteristic indexes, and selecting characteristics with higher scores;
(5) dividing the whole life cycle of the rubber strip into N types of degradation states according to time: learning time domain and frequency domain characteristics of a current curve of the adhesive tape in a full life cycle by using a classification algorithm under a normal state, a degradation state 1, a degradation state 2, a degradation state 3 … …, a degradation state N-2 and a failure state to obtain an adhesive tape degradation state classification model, wherein the normal state, the degradation state 1, the degradation state 2, the degradation state 3 … …, the degradation state N-2 and the failure state are determined according to the service time, and N is more than 4 and less than 10;
(6) and identifying the degradation state of the current finger protection adhesive tape according to the adhesive tape degradation state classification model, calculating the residual life of the adhesive tape, and judging whether the adhesive tape needs to be replaced.
2. The method for predicting the residual life of the finger protection rubber strip of the urban rail door system according to claim 1, characterized in that: in the step (2), the time domain features include a maximum value, a minimum value, a mean value, a variance, a skewness, a kurtosis, and a kurtosis.
3. The method for predicting the residual life of the finger protection rubber strip of the urban rail door system according to claim 1, characterized in that: in the step (2), the extraction method of the frequency domain features is to perform wavelet decomposition on the motor current data to obtain the energy of each sub-band and determine a frequency domain feature set.
4. The method for predicting the residual life of the finger protection rubber strip of the urban rail door system according to claim 1, characterized in that: in the step (3), the feature correlation analysis calculation formula is as follows:
in the formula, ρx,yIs the characteristic correlation, x and y are characteristic values,
the standard deviation is corresponding to the characteristic value.
5. The method for predicting the residual life of the finger protection rubber strip of the urban rail door system according to claim 1, characterized in that: in the step (3), the characteristic sensitivity calculation formula is as follows:
in the formula, λ (x)j) Is xjF (x) is the output value of the classification model, ST is the learning sample, xjIs the characteristic taken.
6. The method for predicting the residual life of the finger protection rubber strip of the urban rail door system according to claim 1, characterized in that: in the step (4), the calculation formula of the Fisher criterion score is as follows:
in the formula (I), the compound is shown in the specification,
is a Fisher-Tropsch discrimination score and,
are respectively a characteristic f1At the average value of the sample space P, Q,
is the corresponding variance.
7. The method for predicting the residual life of the finger protection rubber strip of the urban rail door system according to claim 1, characterized in that: in the step (5), an intelligent algorithm is used for constructing the adhesive tape degradation state classification model, and the method comprises the following steps:
and (4) outputting the N-2 degradation state labels as samples, forming a feature vector with the features selected in the step (4), randomly dividing the feature vector into learning samples and testing samples, inputting the learning samples into an intelligent algorithm for training to obtain an adhesive tape degradation state classification model, verifying the classification accuracy of the classification model by using the testing samples, and if the classification accuracy does not meet the set requirement, adjusting algorithm parameters until the classification accuracy meets the requirement.
8. The method for predicting the residual life of the finger protection rubber strip of the urban rail door system according to claim 1, characterized in that: in the step (6), the remaining life calculation formula is as follows:
wherein i is 1,2iFor the residual service life of the adhesive tape in different states, T (k) is the residual service life of the adhesive tape in the kth state, and k is the current state type of the adhesive tape.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010563092.XA CN113821866B (en) | 2020-06-19 | 2020-06-19 | Method for predicting residual life of finger protection adhesive tape of urban rail door system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010563092.XA CN113821866B (en) | 2020-06-19 | 2020-06-19 | Method for predicting residual life of finger protection adhesive tape of urban rail door system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113821866A true CN113821866A (en) | 2021-12-21 |
| CN113821866B CN113821866B (en) | 2024-01-19 |
Family
ID=78924595
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010563092.XA Active CN113821866B (en) | 2020-06-19 | 2020-06-19 | Method for predicting residual life of finger protection adhesive tape of urban rail door system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113821866B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN206693836U (en) * | 2016-12-09 | 2017-12-01 | 重庆海德世拉索系统（集团）有限公司 | Automobile back door intelligence control system |
| CN107609316A (en) * | 2017-10-25 | 2018-01-19 | 南京康尼机电股份有限公司 | A kind of virtual prototype method for building up of rail traffic vehicles stopping sliding door |
| CN109598048A (en) * | 2018-11-27 | 2019-04-09 | 杭州市地铁集团有限责任公司 | A kind of lubrication degradation prediction technique of track vehicle door system |
| US20200184131A1 (en) * | 2018-06-27 | 2020-06-11 | Dalian University Of Technology | A method for prediction of key performance parameter of an aero-engine transition state acceleration process based on space reconstruction |
-
2020
- 2020-06-19 CN CN202010563092.XA patent/CN113821866B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN206693836U (en) * | 2016-12-09 | 2017-12-01 | 重庆海德世拉索系统（集团）有限公司 | Automobile back door intelligence control system |
| CN107609316A (en) * | 2017-10-25 | 2018-01-19 | 南京康尼机电股份有限公司 | A kind of virtual prototype method for building up of rail traffic vehicles stopping sliding door |
| US20200184131A1 (en) * | 2018-06-27 | 2020-06-11 | Dalian University Of Technology | A method for prediction of key performance parameter of an aero-engine transition state acceleration process based on space reconstruction |
| CN109598048A (en) * | 2018-11-27 | 2019-04-09 | 杭州市地铁集团有限责任公司 | A kind of lubrication degradation prediction technique of track vehicle door system |
Non-Patent Citations (1)
| Title |
|---|
| 杜宏亮;: "车辆雨刮系统原理及故障分析", 汽车与驾驶维修(维修版), no. 07 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113821866B (en) | 2024-01-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109949823B (en) | DWPT-MFCC and GMM-based in-vehicle abnormal sound identification method | |
| CN110405537B (en) | Method for establishing guide rail precision prediction model based on deep learning | |
| CN103810374A (en) | Machine fault prediction method based on MFCC feature extraction | |
| CN106323452A (en) | Equipment abnormal sound detection method and detecting device | |
| CN110259648B (en) | Fan blade fault diagnosis method based on optimized K-means clustering | |
| CN110717379A (en) | Health assessment method for subway car door key components based on feature fusion | |
| CN107844067B (en) | A kind of gate of hydropower station on-line condition monitoring control method and monitoring system | |
| WO2023103268A1 (en) | Pump valve fault detection method | |
| CN111398798B (en) | Circuit breaker energy storage state identification method based on vibration signal interval feature extraction | |
| CN104596780A (en) | Diagnosis method for sensor faults of motor train unit braking system | |
| CN111239672B (en) | Machine learning algorithm-based gradient fault prediction method for optical fiber current transformer | |
| CN116778964A (en) | Power transformation equipment fault monitoring system and method based on voiceprint recognition | |
| CN111353640B (en) | Method for constructing wind speed prediction model by combination method | |
| CN114118219A (en) | Data-driven real-time abnormal detection method for health state of long-term power-on equipment | |
| CN115081331A (en) | Wind turbine generator running state abnormity detection method based on state parameter reconstruction error | |
| CN116776075A (en) | Fan blade health monitoring system and monitoring method thereof | |
| CN110220885B (en) | Comprehensive judgment method for wear state of mechanical equipment | |
| CN109580260A (en) | A kind of inferior health diagnostic method of track vehicle door system | |
| CN112861443B (en) | Advanced learning fault diagnosis method integrated with priori knowledge | |
| CN114330096A (en) | Deep learning-based rail transit motor bearing fault degree classification model construction method | |
| CN109598048A (en) | A kind of lubrication degradation prediction technique of track vehicle door system | |
| CN113821866B (en) | Method for predicting residual life of finger protection adhesive tape of urban rail door system | |
| CN109840386B (en) | Damage identification method based on factor analysis | |
| CN116720073A (en) | Abnormality detection extraction method and system based on classifier | |
| CN103400021A (en) | KDSNE hydroelectric generating set noise source detection method based on heavy tailed distribution |
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 |