CN109443766A - A kind of heavy-duty vehicle gearbox gear Safety Analysis Method - Google Patents
A kind of heavy-duty vehicle gearbox gear Safety Analysis Method Download PDFInfo
- Publication number
- CN109443766A CN109443766A CN201811051014.0A CN201811051014A CN109443766A CN 109443766 A CN109443766 A CN 109443766A CN 201811051014 A CN201811051014 A CN 201811051014A CN 109443766 A CN109443766 A CN 109443766A
- Authority
- CN
- China
- Prior art keywords
- index
- heavy
- rule
- duty vehicle
- vehicle gearbox
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/02—Gearings; Transmission mechanisms
- G01M13/028—Acoustic or vibration analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
Abstract
The present invention provides a kind of heavy-duty vehicle gearbox gear Safety Analysis Method, it is characterized by: real-time monitoring and acquire tested heavy-duty vehicle gearbox gear vibration signal, the optimal index subset of maximum correlation and minimum redundancy is selected by sensitive factor based on distance and maximum correlation coefficient (MIC)；Simultaneously according to the representative index of selection, establish the model for being based on confidence rule base (BRB) and ER reasoning, dimensionality reduction and decorrelation are carried out to the feature vector of heavy-duty vehicle gearbox running state parameter composition, thus the high precision analysis of the heavy-duty vehicle gearbox safe condition obtained.The invention has the advantages that (1) can carry out dimensionality reduction and decorrelation to the feature vector being made of heavy-duty vehicle gearbox gear running state parameter；(2) precision of model is improved；(3) this method has important economic benefit by the assessment to gear condition, the maintenance of accurate judgement transmission gear, the period for repairing and changing part；(4) the burst probability and downtime that can reduce accident for improving vehicle driving safety there is important theory and practice to be worth.
Description
Technical field
The invention belongs to heavy-duty vehicle security of system assessment technology field, in particular to a kind of heavy-duty vehicle system is examined
Consider index related and redundancy Safety Analysis Method.
Background technique
Gearbox is the principal assembly of heavy-duty vehicle power train, is responsible for power being transmitted to driving system from engine
Important mission.According to statistics, 60% failure is up in the failure of gearbox as gear class failure.Heavy-duty vehicle transmission gear
Because of the limitation of its functional requirement, determines that transmission gear must satisfy to run at high speed and transmitted with large torque, make it compared to general
Gear facewidth under the premise of finite lifetime is thinner, and transimission power density is bigger, and abrasion is more serious, and reliability service is with regard to outstanding
It is important.
The operating status judgement of heavy-duty vehicle gearbox relies primarily on the mode of mutual disassembly, observation at present, using regular
It overhauls with the method for correction maintenance and ensures its operation, detection efficiency is lower, or even implies biggish security risk.In addition,
The platform experiment of heavy-duty vehicle gearbox unit time is costly, does not allow to carry out large sample, prolonged platform experiment,
Traditional reliability estimation method being distributed based on fault statistics data and life index is simultaneously not suitable for.
Existing in gear-driven security assessment method, Lei Yaguo from vibration signal by extracting time domain
And frequency-domain index carries out gear-box state with k-nearest neighbor using the feature evaluation choice of technology index of correlation based on distance
Identification.Wang Zhala is predicted and is analyzed to gearbox abrasion using gray system theory and fractal theory.Liu Pei is with oil liquid
Medium selects feature using principle component analysis, and gray theory and artificial neural network is respectively adopted to gear wear shape
State is predicted.Feature vector is extracted in Cai Quancong vibration signal, dimension-reduction treatment is carried out to numerous features with pivot analysis, and
Gearbox fault is identified using depth convolutional neural networks model.The drawbacks of existing appraisal procedure, is: index selection
The correlation especially redundancy between index is not accounted in the process.Uncorrelated and redundancy index will increase calculation amount, reduce and comment
Estimate precision.In addition, each method cannot merge qualitative, quantitative information, interpretable, trackability as a result is not strong.
Summary of the invention
The purpose of the present invention is: the vibration signal of gearbox gear is monitored by sensor, and then is mentioned from vibration signal
A variety of time domain indexes are taken, related and not redundancy ideal indicator collection is selected to maximum information coefficient (MIC) by sensitive factor；
Confidence rule base Security Evaluation Model is established, real-time monitoring and effective security evaluation are carried out to heavy-duty vehicle gearbox gear.
Insight of the invention is that for the state status of heavy-duty vehicle gearbox gear different aspect reaction, real-time monitoring
And the state parameter of tested heavy-duty vehicle gearbox gear is acquired, pass through sensitive factor and maximum correlation coefficient based on distance
(MIC) the optimal index subset with assessment result maximum correlation and minimum redundancy is selected；Simultaneously according to the representative of selection
Property index, establish be based on confidence rule base (BRB) and ER reasoning model, to heavy-duty vehicle gearbox running state parameter group
At feature vector carry out dimensionality reduction and decorrelation, thus the high precision analysis of the heavy-duty vehicle gearbox safe condition obtained.
The present invention provides a kind of heavy-duty vehicle gearbox gear Safety Analysis Method, it is characterised in that: real-time monitoring is simultaneously
Tested heavy-duty vehicle gearbox gear vibration signal is acquired, sensitive factor and maximum correlation coefficient (MIC) based on distance are passed through
Select the optimal index subset of maximum correlation and minimum redundancy；Simultaneously according to the representative index of selection, foundation is based on
The model of confidence rule base (BRB) and ER reasoning carry out the feature vector of heavy-duty vehicle gearbox running state parameter composition
Dimensionality reduction and decorrelation, thus the high precision analysis of the heavy-duty vehicle gearbox safe condition obtained, specifically includes the following steps:
Step 1: the acquisition and processing of heavy-duty vehicle gearbox gear vibration signal
If heavy-duty vehicle gearbox gear has three state, normal value (N), medium outage (gear crack M), catastrophe failure
(gear peels off S).Firstly, heavy-duty vehicle gearbox gear several groups real-time monitoring vibration signal is extracted, then from every group of vibration
9 characteristic variables sensitive to the system failure, that diagnosis can be directly used in, the mathematical description and object of each index are extracted in signal
It is as follows to manage meaning:
Mean value (M) u=E [x (t)]
Root-mean-square value (RMS)
Variance (V) σ2=E [(x (t)-μ)2]
Flexure (S) α=E [(x (t)-μ)3]/σ3
Kurtosis (K) β=E [(x (t)-μ)4]/σ4
Peak index (CF)
Pulse index (IF)
Waveform index (SF)
Margin index (CLF)
Step 2: the optimal index selection of heavy-duty vehicle gearbox gear assessment
A large amount of index is extracted from vibration signal to react the state of gear, wherein uncorrelated and redundancy index meeting
Increase calculation amount and reduce Evaluation accuracy, therefore passes through a kind of two stages feature for mutually separating feature correlation with redundancy
Selection method selects the optimal index collection with maximum correlation and minimum redundancy；
Step 2.1 is based on Sensitivity Factor and selects index of correlation
Susceptibility sequence is carried out to feature using the feature evaluation technology based on distance, deletes phase unrelated or weak with output
The feature of pass.It assesses principle: characteristic distance is bigger between the class of a certain index, characteristic distance is smaller in class, then the index is got over
It is sensitive, higher with results relevance.The index that sensitive factor is less than given threshold value is uncorrelated index, needs to delete；
(1) average distance between same class sample is calculated
μ in formulaiClass ωiSample average, c are the classification number of sample；
(2) calculate inhomogeneity between sample average distance
(3) Sensitivity Factor is calculated
E=db/dw (3)
(4) Sensitivity Factor normalizes
The Sensitivity Factor of j-th of feature is calculated by formula (1)~(3), and then Sensitivity Factor is returned
One changes:
Step 2.2 is based on maximum correlation coefficient and deletes redundancy index
The redundancy feature is analyzed according to maximum information coefficient (MIC), evaluation criterion are as follows: between two indexes
Maximum information coefficient (MIC) it is bigger, the redundancy between index is more；Maximum information coefficient (MIC) is greater than given threshold value
When, delete the small index of sensitive factor；
Maximum information coefficient is mainly calculated using mutual information and Meshing Method；Assuming that there is sequence D={ (ai,
bi), i=1,2 ..., n }.Mutual information between variables A and B is defined as:
In formula, p (a, b) is joint probability density, and p (a) and p (b) are marginal probability density；A and B is drawn at random respectively
It is divided into x sections and y sections；
Maximum correlation coefficient (MIC) is defined as:
B (n) is the upper limit value of grid dividing x × y, generally, B (n)=n in formula0.6；
If MIC (xi,xj) < K1, it is believed that index xiWith index xjIndependently of each other；If MIC (fi,fj)≥K1, then think
Index xiWith index xjCorrelation, the lesser index of weight will be deleted.Wherein K1It is practical according to engineering for preset threshold value
It formulates；
Step 3: the building of heavy-duty vehicle gearbox gear assessment models
According to the representative index of selection, the assessment for being based on confidence rule base (BRB) and evidence (ER) reasoning algorithm is established
Model；
Step 3.1 confidence rule base model
Security evaluation is carried out using confidence rule base model, wherein kth confidence rule is described as follows:
In formula (6),Indicate i-th of premise attribute x in kth ruleiGinseng
Examine value；L indicates number regular in total confidence rule base；And Ai={ Ai,j, j=1 ..., MiIndicate by i-th
The M of a premise attributeiSet composed by a reference value； θk(k=1 ..., L) indicates the regular weight of kth rule, it is anti-
Different degree of the kth rule relative to Else Rule in confidence rule base is reflected；δik(i=1 ..., M；K=1 ..., L) table
Show the weight of i-th of premise attribute in kth rule, it reflects i-th of premise attribute relative to other premise attributes
Different degree； βj,k(j=1 ..., N, k=1 ..., L) indicate in kth rule relative to output par, c (i.e. confidence advise
The part Then then) j-th of assessment resultConfidence level；The parameter of confidence rule base is needed according to expertise and warp
Test determination；
Step 3.2 evidence (ER) reasoning
BRB rule base needs to obtain last assessment result by ER reasoning；One first input can will BRB rule in
Respective rule activation.Activate the weighing computation method of rule are as follows:
Wherein ωk∈ [0,1], k=1 ..., L；L is the number of activation rule.It is opposite to input
The confidence level of premise attribute in the rule of correspondence；θkFor regular weight；δiFor attribute weight.
The final output of BRB rule base polymerize corresponding rule by ER reasoning algorithm to realize.
Advantage of the present invention:
(1) dimensionality reduction can be carried out to the feature vector being made of heavy-duty vehicle gearbox gear running state parameter and is gone
It is related；
(2) precision of model is improved；
(3) this method is by the assessment to gear condition, the maintenance of accurate judgement transmission gear, the period for repairing and changing part,
With important economic benefit；
(4) the burst probability and downtime that can reduce accident have important reason for improving vehicle driving safety
By and more practical value.
Detailed description of the invention
Fig. 1 security state evaluation process
Each index sensitive factor of Fig. 2
Fig. 3 uses the heavy-duty vehicle gearbox state assessment result of confidence rule base (BRB)
Specific implementation method
The process of this method is as shown in Figure 1, mainly comprise the steps that
Step 1: the acquisition and processing of heavy-duty vehicle gearbox gear vibration signal
Two sensors are placed close to the place of gear in gear box.Acceleration transducer has used the U.S.
Dytran 3023M2 (three-dimensional) universal acceleration transducer, sensitivity are ± 10mv/g.Each sensor respectively measure it is horizontal,
Vertical and vertical direction vibration signal.Data sampling frequency is 25600Hz.There are three types of working gear state is total: first is that normal
Working condition；Second is that gear crack；Third is that gear peels off.
Measure 6 groups of vibration signals in the present invention altogether, the respectively x, y, z that measures of sensor 1 and 2 is to signal.Every group of vibration
9 time domain indexes of signal extraction, thus share 54 time domain indexes never ipsilateral reflection gearbox gear state.
Step 2: the optimal index selection of heavy-duty vehicle gearbox gear assessment
Step 2.1 is based on Sensitivity Factor and selects index of correlation
Sensitive factor of each index relative to gearbox gear state is calculated by formula (1)~(4).Each index is sensitive
Factor calculated result is as shown in Figure 2.Sensitive factor threshold value φ is set as 0.5, and index of the sensitive factor less than 0.5 is non-correlation
Index needs to delete.Accordingly, the root mean square of the root mean square (RMS-4) of signal 4, the root mean square (RMS-2) of signal 2, signal 1
(RMS-1), the pulse (IF-2) of the variance (V-4) of signal 4, the peak value (CF-2) of signal 2 and signal 2 is chosen as index of correlation.
Step 2.2 is based on maximum correlation coefficient and deletes redundancy index
For selected index of correlation, the maximum correlation coefficient (MIC) between two indexes is calculated separately.Maximal correlation is set
Coefficient threshold is 0.5.The information weight for including between being considered two indexes if the maximum correlation coefficient between two indexes is greater than 0.5
Multiple, the small index of sensitive factor is redundancy index in two indexes, needs to be deleted.
The maximum correlation coefficient that 3 index of table is asked
So far, RMS-4, RMS-2 and CF-2 are chosen as the optimal index with maximum correlation and minimum redundancy.
Step 3: the building of heavy-duty vehicle gearbox gear assessment models
Selected optimal index realizes gearbox as the input in confidence rule base, according to the confidence rule base of foundation
Status assessment.
Step 3.1 confidence rule base model
(1) premise attribute reference value is determined
The reference value of premise attribute is the guarantee of confidence rule base scale and precision.The determination of premise attribute is usually root
It is determined according to expertise.
The definition of 4 RMS-4 reference value of table
The definition of 5 RMS-2 reference value of table
The definition of 6 CF-2 reference value of table
Gear has three state, normal value (N), medium outage (gear crack M), catastrophe failure (gear peels off S).Wherein
Gearbox in medium outage, which needs to maintain, can just continue to use, and the gearbox in catastrophe failure then needs replacing or
It scraps.
The definition of 7 gearbox gear operating mode of table
(2) input information conversion
Using rule-based (rule) or the input information method for transformation of effectiveness (utility), x will be inputtediIt is converted into phase
For reference value Ai,j(j=1 ..., Ji) confidence alphai,j, wherein reference value is ascending order arrangement.αi,jCalculation method such as
Under:
αi,j+1=1- αi,j if γi,j≤xi≤γi,j+1, j=1 ..., Ji-1 (11)
αi,s=0 for s=1 ..., Ji,s≠j,j+1 (12)
For example, RMS-4 value is 5.4326, then the confidence level relative to reference value is inputted are as follows:
{ (L, 0), (M, 0.5429), (H, 0.4571), (VH, 0) } wherein 0.5429=(6.4967-5.4326)/
(6.4967-4.5366).Remainder data is converted in this manner.
(3) the confidence rule Kuku (BRB) for being directed to heavy-duty vehicle gearbox gear security evaluation is established
In order to realize the security evaluation of heavy-duty vehicle gearbox gear, the confidence rule base expert system of science is constructed extremely
It closes important.In confidence rule base, it is desirable that traverse all premise attribute reference values.Therefore need to establish 4 × 4 × 3 totally 48
Confidence rule is then.Heavy-duty vehicle gearbox gear security evaluation confidence rule as shown in table 3 is constructed according to expertise
Library.
By taking the 2nd rule in confidence rule base as an example, description are as follows:
R2:If(x1 is M)∧(x2 is L)∧(x3 is L),Then{(N,0.6421)(M,0.1895),(S,
0.1684)}
with rule weight 0.6303.
Its meaning are as follows: if it is L that RMS-4, which is M, RMS-2 L, and CF-2, then heavy-duty vehicle gearbox gear is in just
The confidence level of normal state is 0.6421, and the confidence level in medium outage is 0.1895, and the confidence level in catastrophe failure is
0.1684.The weight of this rule is 0.6303.
The confidence rule base of 8 heavy-duty vehicle gearbox gear status assessment of table
Step 3.2 evidence (ER) reasoning
It inputs after the completion of information conversion, several rules in confidence rule base will be activated and by ER inference method
Aggregate into final output.According to formula (7)~(9), the security state evaluation of heavy-duty vehicle gearbox gear is realized, such as
Shown in Fig. 3, assessment accuracy rate is 93.33%.
It is close with support vector machines (SVM), BP neural network, fuzzy reasoning and K in order to prove the validity of proposed method
Adjacent method compares research.For other methods, this method have highest precision, minimum minimum variance (MSE) and
Strongest robustness.In addition, the modeling of BRB and reasoning process are high-visible, parameter is easy to adjust, as a result have trackability,
More easily explain.
9 heavy-duty vehicle gearbox gear status safety of table assesses distinct methods Comparative result
Claims (1)
1. a kind of heavy-duty vehicle gearbox gear Safety Analysis Method, it is characterised in that: real-time monitoring simultaneously acquires tested heavy duty
Vehicle transmission gear vibration signal selects maximal correlation by sensitive factor based on distance and maximum correlation coefficient MIC
The optimal index subset of property and minimum redundancy；Simultaneously according to the representative index of selection, establish based on confidence rule base BRB and
The model of ER reasoning carries out dimensionality reduction and decorrelation to the feature vector of heavy-duty vehicle gearbox running state parameter composition, thus
The high precision of the heavy-duty vehicle gearbox safe condition of acquisition is analyzed, specifically includes the following steps:
Step 1: the acquisition and processing of heavy-duty vehicle gearbox gear vibration signal
If heavy-duty vehicle gearbox gear has three state, normal value N, medium outage, that is, gear crack M, catastrophe failure, that is, gear
Peel off S；Firstly, heavy-duty vehicle gearbox gear several groups real-time monitoring vibration signal is extracted, then from every group of vibration signal
9 characteristic variables sensitive to the system failure, that diagnosis can be directly used in are extracted, the mathematical description and physical significance of each index are such as
Under:
Mean value M u=E [x (t)]
Root-mean-square value RMS
Variance V σ2=E [(x (t)-μ)2]
Flexure S α=E [(x (t)-μ)3]/σ3
Kurtosis K β=E [(x (t)-μ)4]/σ4
Peak index CF
Pulse index IF
Waveform index S F
Margin index CLF
Step 2: the optimal index selection of heavy-duty vehicle gearbox gear assessment
A large amount of index is extracted from vibration signal to react the state of gear, wherein uncorrelated and redundancy index will increase meter
Calculation amount and reduction Evaluation accuracy, therefore pass through a kind of two stages feature selection approach for mutually separating feature correlation with redundancy
Select the optimal index collection with maximum correlation and minimum redundancy；
Step 2.1 is based on Sensitivity Factor and selects index of correlation
Susceptibility sequence is carried out to feature using the feature evaluation technology based on distance, delete and exports unrelated or weak relevant spy
Sign.It assesses principle: characteristic distance is bigger between the class of a certain index, characteristic distance is smaller in class, then the index it is more sensitive, with
Results relevance is higher；The index that sensitive factor is less than given threshold value is uncorrelated index, needs to delete；
(1) average distance between same class sample is calculated
μ in formulaiClass ωiSample average, c are the classification number of sample；
(2) calculate inhomogeneity between sample average distance
(3) Sensitivity Factor is calculated
E=db/dw (3)
(4) Sensitivity Factor normalizes
The Sensitivity Factor of j-th of feature is calculated by formula (1)~(3), and then Sensitivity Factor is normalized:
Step 2.2 is based on maximum correlation coefficient and deletes redundancy index
The redundancy feature is analyzed according to maximum information coefficient MIC, evaluation criterion are as follows: the maximum letter between two indexes
Breath coefficient MIC is bigger, and the redundancy between index is more；When maximum information coefficient MIC is greater than given threshold value, sensitive factor is deleted
Small index；
Maximum information coefficient is mainly calculated using mutual information and Meshing Method；Assuming that there is sequence D={ (ai,bi), i=
1,2,...,n}.Mutual information between variables A and B is defined as:
In formula, p (a, b) is joint probability density, and p (a) and p (b) are marginal probability density；A and B is respectively x by random division
Section and y sections；
Maximum correlation coefficient MIC is defined as:
B (n) is the upper limit value of grid dividing x × y, generally, B (n)=n in formula0.6；
If MIC (xi,xj) < K1, it is believed that index xiWith index xjIndependently of each other；If MIC (fi,fj)≥K1, then think index
xiWith index xjCorrelation, the lesser index of weight will be deleted；Wherein K1For preset threshold value, formulated according to engineering is practical；
Step 3: the building of heavy-duty vehicle gearbox gear assessment models
According to the representative index of selection, the assessment models based on confidence rule base BRB and evidence ER reasoning algorithm are established；
Step 3.1 confidence rule base model
Security evaluation is carried out using confidence rule base model, wherein kth confidence rule is described as follows:
With a rule weightθk and attribute weightδ1,k,δ2,k,...,δ6,k (6)
In formula (6),Indicate i-th of premise attribute x in kth ruleiReference value；L
Indicate number regular in total confidence rule base；And Ai={ Ai,j, j=1 ..., MiIndicate by i-th of premise
The M of attributeiSet composed by a reference value；θk(k=1 ..., L) indicates the regular weight of kth rule, it reflects kth
Different degree of the rule relative to Else Rule in confidence rule base；δik(i=1 ..., M；K=1 ..., L) it indicates in kth item
The weight of i-th of premise attribute in rule, it reflects different degree of i-th of premise attribute relative to other premise attributes；βj,k
(j=1 ..., N, k=1 ..., L) it indicates in kth rule relative to (the i.e. portion Then of confidence rule of output par, c
Point) j-th of assessment resultConfidence level；The parameter of confidence rule base is needed according to expertise and empirically determined；
Step 3.2 evidence ER reasoning
BRB rule base needs to obtain last assessment result by ER reasoning；One first input can will BRB rule in it is corresponding
Rale activation；Activate the weighing computation method of rule are as follows:
Wherein ωk∈ [0,1], k=1 ..., L；L is the number of activation rule；To input relative to correspondence
The confidence level of premise attribute in rule；θkFor regular weight；δiFor attribute weight；
The final output of BRB rule base polymerize corresponding rule by ER reasoning algorithm to realize；
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811051014.0A CN109443766A (en) | 2018-09-10 | 2018-09-10 | A kind of heavy-duty vehicle gearbox gear Safety Analysis Method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811051014.0A CN109443766A (en) | 2018-09-10 | 2018-09-10 | A kind of heavy-duty vehicle gearbox gear Safety Analysis Method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109443766A true CN109443766A (en) | 2019-03-08 |
Family
ID=65530943
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811051014.0A Pending CN109443766A (en) | 2018-09-10 | 2018-09-10 | A kind of heavy-duty vehicle gearbox gear Safety Analysis Method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109443766A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110056640A (en) * | 2019-04-12 | 2019-07-26 | 苏芯物联技术(南京)有限公司 | Speed reducer wireless malfunction diagnostic method based on acceleration signal and edge calculations |
| CN110348752A (en) * | 2019-07-16 | 2019-10-18 | 中国人民解放军火箭军工程大学 | A kind of large scale industry system structure security assessment method considering environmental disturbances |
| CN110503813A (en) * | 2019-08-28 | 2019-11-26 | 北京无线电计量测试研究所 | A kind of anti-excavation monitoring method for early warning of buried pipeline |
| CN112396177A (en) * | 2019-08-12 | 2021-02-23 | 美光科技公司 | Predictive repair of vehicles |
| CN112861383A (en) * | 2021-03-17 | 2021-05-28 | 哈尔滨工业大学 | Railway station anti-seismic toughness evaluation method and system |
| CN115755608A (en) * | 2022-11-18 | 2023-03-07 | 沈阳盛世五寰科技有限公司 | Energy consumption optimization decision method for high-pressure roller mill |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105139086A (en) * | 2015-08-13 | 2015-12-09 | 杭州电子科技大学 | Track profile irregularity amplitude estimation method employing optimal belief rules based inference |
| CN105975589A (en) * | 2016-05-06 | 2016-09-28 | 哈尔滨理工大学 | Feature selection method and device of high-dimension data |
| CN106059412A (en) * | 2016-01-28 | 2016-10-26 | 杭州电子科技大学 | Method for controlling rotating speed of separately excited DC motor based on belief rule base reasoning |
| CN107727395A (en) * | 2017-07-21 | 2018-02-23 | 中国矿业大学 | A kind of Method for Bearing Fault Diagnosis assessed based on full variation and uncompensation distance |
-
2018
- 2018-09-10 CN CN201811051014.0A patent/CN109443766A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105139086A (en) * | 2015-08-13 | 2015-12-09 | 杭州电子科技大学 | Track profile irregularity amplitude estimation method employing optimal belief rules based inference |
| CN106059412A (en) * | 2016-01-28 | 2016-10-26 | 杭州电子科技大学 | Method for controlling rotating speed of separately excited DC motor based on belief rule base reasoning |
| CN105975589A (en) * | 2016-05-06 | 2016-09-28 | 哈尔滨理工大学 | Feature selection method and device of high-dimension data |
| CN107727395A (en) * | 2017-07-21 | 2018-02-23 | 中国矿业大学 | A kind of Method for Bearing Fault Diagnosis assessed based on full variation and uncompensation distance |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110056640A (en) * | 2019-04-12 | 2019-07-26 | 苏芯物联技术(南京)有限公司 | Speed reducer wireless malfunction diagnostic method based on acceleration signal and edge calculations |
| CN110056640B (en) * | 2019-04-12 | 2021-07-06 | 苏芯物联技术(南京)有限公司 | Speed reducer wireless fault diagnosis method based on acceleration signal and edge calculation |
| CN110348752A (en) * | 2019-07-16 | 2019-10-18 | 中国人民解放军火箭军工程大学 | A kind of large scale industry system structure security assessment method considering environmental disturbances |
| CN110348752B (en) * | 2019-07-16 | 2023-07-25 | 中国人民解放军火箭军工程大学 | Large industrial system structure safety assessment method considering environmental interference |
| CN112396177A (en) * | 2019-08-12 | 2021-02-23 | 美光科技公司 | Predictive repair of vehicles |
| CN110503813A (en) * | 2019-08-28 | 2019-11-26 | 北京无线电计量测试研究所 | A kind of anti-excavation monitoring method for early warning of buried pipeline |
| CN110503813B (en) * | 2019-08-28 | 2020-11-24 | 北京无线电计量测试研究所 | Anti-excavation monitoring and early warning method for buried pipeline |
| CN112861383A (en) * | 2021-03-17 | 2021-05-28 | 哈尔滨工业大学 | Railway station anti-seismic toughness evaluation method and system |
| CN112861383B (en) * | 2021-03-17 | 2022-09-16 | 哈尔滨工业大学 | Railway station anti-seismic toughness evaluation method and system |
| CN115755608A (en) * | 2022-11-18 | 2023-03-07 | 沈阳盛世五寰科技有限公司 | Energy consumption optimization decision method for high-pressure roller mill |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109443766A (en) | A kind of heavy-duty vehicle gearbox gear Safety Analysis Method | |
| CN106006344B (en) | Staircase On-line Fault early warning system and method for diagnosing faults | |
| CN106769052B (en) | A kind of mechanical system rolling bearing intelligent failure diagnosis method based on clustering | |
| CN105241660B (en) | High-speed rail large bridge performance test methods based on health monitoring data | |
| Bafroui et al. | Application of wavelet energy and Shannon entropy for feature extraction in gearbox fault detection under varying speed conditions | |
| CN109974782B (en) | Equipment fault early warning method and system based on big data sensitive characteristic optimization selection | |
| CN101660969B (en) | Intelligent fault diagnosis method for gear box | |
| CN104712542B (en) | A kind of reciprocating compressor sensitive features based on Internet of Things are extracted and method for diagnosing faults | |
| CN111830408A (en) | Motor fault diagnosis system and method based on edge calculation and deep learning | |
| CN100470416C (en) | Power plant thermal equipment intelligent state diagnosing and analyzing system | |
| CN108256556A (en) | Wind-driven generator group wheel box method for diagnosing faults based on depth belief network | |
| CN103291600B (en) | Fault diagnosis method for hydraulic pump based on EMD-AR (empirical mode decomposition-auto-regressive) and MTS (mahalanobis taguchi system) | |
| CN109657880A (en) | A kind of Collision Accidents of Ships grade prediction technique and system based on Bayesian network | |
| CN110175369A (en) | A kind of gear method for predicting residual useful life based on two-dimensional convolution neural network | |
| CN102736562B (en) | Knowledge base construction method oriented to fault diagnosis and fault prediction of numerical control machine tool | |
| CN102155988A (en) | Equipment monitoring and diagnosing method | |
| CN114894468B (en) | Chaos detection-based early weak fault diagnosis method for planetary gear box | |
| CN111723449B (en) | Performance degradation evaluation method for constant-deceleration braking system of mine hoist | |
| CN108982096A (en) | Industrial robot crank axle wear detecting method based on heuristic rule system | |
| Cao et al. | Deterioration state diagnosis and wear evolution evaluation of planetary gearbox using vibration and wear debris analysis | |
| CN113092112A (en) | Bearing composite fault diagnosis method based on EEMD multi-feature fusion | |
| CN116595319A (en) | Prediction method and system applied to rail transit motor health state evaluation | |
| Cao et al. | A novel spiking graph attention network for intelligent fault diagnosis of planetary gearboxes | |
| CN117115637A (en) | Water quality monitoring and early warning method and system based on big data technology | |
| CN113973403B (en) | Temperature-induced strain field redistribution intelligent sensing method based on structure discrete measurement point topology |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190308 |