CN109000921A - A kind of diagnostic method of wind generator set main shaft failure - Google Patents
A kind of diagnostic method of wind generator set main shaft failure Download PDFInfo
- Publication number
- CN109000921A CN109000921A CN201710423126.3A CN201710423126A CN109000921A CN 109000921 A CN109000921 A CN 109000921A CN 201710423126 A CN201710423126 A CN 201710423126A CN 109000921 A CN109000921 A CN 109000921A
- Authority
- CN
- China
- Prior art keywords
- model
- signal
- modal components
- diagnostic method
- fault
- 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
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
Abstract
The invention proposes a kind of diagnostic methods of wind generator set main shaft failure, including vibration signals collecting, VMD variation mode decomposition, sensitive factor assessment, signal reconstruction, VPMCD fault diagnosis, by carrying out VMD analysis to fan vibration signal, by signal decomposition at the modal components of one group of different frequency, since fault signature generally only occurs at specific frequency range, therefore analysis and assessment are carried out to modal components using sensitive factor, screening includes the component of fault signature, exclude noise jamming component, and then reconstruction signal, highlight fault message；Then the characteristic value constitutive characteristic for extracting reconstruction signal identifies vector, carries out fault diagnosis by establishing characteristic value internal relation in conjunction with VPMCD method, avoids the selection of traditional classifier subjective parameters and searching process, shorten failure diagnosis time.
Description
Technical field
The present invention relates to fault diagnosis technology field more particularly to wind generator set main shaft fault diagnosises, and in particular to a kind of
The diagnostic method of wind generator set main shaft failure.
Background technique
As China develops emphatically the non-fossil energy, the installed capacity of wind-power electricity generation increases year by year.Wherein, directly-driven wind
Generator gradually becomes main trend because of the advantages that generating efficiency is high, maintenance cost is low, power grid access performance is excellent.However, by
In wind power plant multidigit in adverse circumstances, operating condition is complicated, causes unit operation component to be easily damaged, especially event occurs for main shaft
Barrier will directly cause stopping accident, bring about great losses if failing timely discovery processing.Therefore, many scholars couple in recent years
Blower fan main shaft fault diagnosis is furtherd investigate, and abundant achievement is achieved.Some utilizes empirical mode decomposition (empirical
Mode decomposition, EMD) realize Wind turbines Rolling Bearing Fault Character extraction；Some decomposes local mean value
(local mean decomposition, LMD) and approximate entropy combine, and classify to rolling bearing fault type；Some needles
To the modal overlap problem of LMD method, adaptive high frequency harmonic wave is added, successfully extracts the imbalance fault feature of shafting.EMD and
LMD method belongs to recursive schema decomposition, and there are end effects for this resolution model, and leading to envelope, deformation occurs, causes to decompose
Inaccuracy, become the limitation of such method analysis processing signal.
Summary of the invention
In order to solve the above technical problems, the present invention provides a kind of diagnostic method of wind generator set main shaft failure, mesh
Be, a kind of diagnostic method for capableing of automatic identification main shaft failure classification is provided, it is inaccurate to solve classification in failure diagnostic process
True problem.
The present invention provides a kind of diagnostic method of wind generator set main shaft failure, comprising:
Step 1, vibration signals collecting:
Vibration signal is obtained by the sensor being mounted on wind generator set main shaft seat；
Step 2, VMD variation mode decomposition:
VMD variation mode decomposition is carried out to the vibration signal, obtains the modal components under multiple and different frequencies；
Step 3, sensitive factor is assessed:
The sensitive factor for calculating each modal components, assessing each modal components includes fault signature degree；
Step 4, signal reconstruction:
Cancelling noise interference, reconstruction signal calculate the feature vector of each reconstruction signal, construct model analysis vector；
Step 5, VPMCD fault diagnosis:
Fault identification is carried out to the model analysis vector with prediction model, calculates the error sum of squares of main shaft state VPM,
It is to determine to carry out Classification and Identification with minimum value；
Wherein, VMD is variation mode decomposition, and VPMCD is the pattern-recognition of variable prediction model, and the main shaft state includes
Normally, outer ring failure, inner ring failure, rolling element failure.
As further improvement of the invention, sensor described in step 1 is acceleration transducer.
As further improvement of the invention, step 2 is specifically included:
Step 201, the vibration signal is decomposed as K simple component mode function, the simple component mode function is limited
Band Intrinsic mode function:
Wherein, f (t) is vibration signal, ukIt (t) is k-th of limit band Intrinsic mode function, AkIt (t) is envelope,
For phase function, t is the time；
Step 202, the frequency bandwidth for estimating each simple component mode function, establishes variational methods model:
Wherein, wkFor k-th of simple component mode function uk(t) center angular frequency, { uk}={ u1……uK, { wk}=
{w1……wK,For the adduction of all simple component mode functions；
Step 203, secondary penalty factor a and Lagrangian r is introduced into variational methods model, it is bright obtains augmentation glug
Day equation；
Wherein, a is secondary penalty factor, r Lagrangian；
Step 204, u is initializedk、wkWith r numerical value, is updated using the continuous iteration of alternating direction multipliers method, look for iteration optimization
The saddle point of augmentation Lagrange, determines judgement precision e in sequence1, according to formula is determined, for meeting the feelings of the judgement formula
Condition stops iteration and exports K modal components.
As further improvement of the invention, the judgement formula specifically:
Wherein, e is taken1=1 × 10-6。
As further improvement of the invention, step 3 specifically:
Calculate the sensitive factor λ of each modal components obtained in step 2k,
Wherein, δk=βk- αk, wherein αkFor the related coefficient of modal components and fault-signal, βkFor modal components and normally
The related coefficient of signal.
As further improvement of the invention, step 4 is specifically included:
Step 401, using sort method to the sensitive factor λkIt is descending to be ranked up, obtain new sensitive factor
Sequence λ 'k；
Step 402, according to sensitive factor sequence λ 'k, calculating difference spectrum assessment vector sequence dk,
dk=λ 'k+1-λ′k；
Step 403, the difference spectrum assessment vector sequence d is found outkMaximum value dj；
Step 404, pass through preceding j dkCorresponding sensitive factor λ 'k, corresponding j modal components are found out, are superimposed the j
Modal components obtain reconstruction signal；
Step 405, according to reconstruction signal, feature vector is obtained using singular value decomposition method.
As further improvement of the invention, sort method described in step 401 is bubbling method or cocktail ranking method
Either radix sorting or Shell sorting.
As further improvement of the invention, step 5 is specifically included:
Step 501, feature vector is substituted into variable prediction model, the variable prediction model is VPMi:
Xi=f (Xj,b0,bj,bjj,bjk)+e2,
Wherein, b0,bj,bjj,bjkFor model parameter；e2To predict error；
Step 502, the error sum of squares of each state VPM is calculated, is to determine to carry out Classification and Identification with minimum value.
As further improvement of the invention, the model parameter is obtained using the training of each failure classes vibration data, institute
Stating prediction error is to preset.
As further improvement of the invention, the prediction model is linear model or linear reciprocal model or secondary model
Or secondary interaction models；
The linear model are as follows:
The linear reciprocal model are as follows:
The secondary model are as follows:
The secondary interaction models are as follows:
The invention has the following beneficial effects: by fan vibration signal carry out VMD analysis, by signal decomposition at one group
The modal components of different frequency, since fault signature generally only occurs at specific frequency range, using sensitive factor to mode
Component carries out analysis and assessment, and screening includes the component of fault signature, excludes noise jamming component, and then reconstruction signal, highlights event
Hinder information；Then the characteristic value constitutive characteristic for extracting reconstruction signal identifies vector, in conjunction with VPMCD method by establishing feature vector
Internal relation carries out fault diagnosis, the selection of traditional classifier subjective parameters and searching process is avoided, when shortening fault diagnosis
Between.
Detailed description of the invention
Fig. 1 is a kind of flow chart of the diagnostic method of wind generator set main shaft failure in first embodiment of the invention；
Fig. 2 is the arrangement of sensitive factor sequence descending and sensitive factor difference spectrogram in first embodiment of the invention；
Fig. 3 is the model analysis vector tendency chart of different conditions sample of signal in first embodiment of the invention.
Specific embodiment
With reference to the accompanying drawing and specific embodiment is described in further detail the present invention.
Embodiment 1, as shown in Figure 1, a kind of diagnostic method of wind generator set main shaft failure, comprising:
Step 1, vibration signals collecting:
Vibration signal is obtained by the sensor being mounted on wind generator set main shaft seat；
Sensor is acceleration transducer, and sample frequency 10kHz, sampling number is 1000 points.
Step 2, VMD variation mode decomposition:
VMD variation mode decomposition is carried out to vibration signal, obtains the modal components under multiple and different frequencies；
Wherein, VMD is variation mode decomposition；
It specifically includes:
Step 201, the vibration signal is decomposed as K simple component mode function, the simple component mode function is limited
Band Intrinsic mode function:
Wherein, f (t) is vibration signal, ukIt (t) is k-th of limit band Intrinsic mode function, AkIt (t) is envelope,
For phase function, t is the time；
Step 202, the frequency bandwidth for estimating each simple component mode function, establishes variational methods model:
Wherein, wkFor k-th of simple component mode function uk(t) center angular frequency, { uk}={ u1……uK, { wk}=
{w1……wK,For the adduction of all simple component mode functions, δ (t) is unit pulse signal；
Step 203, secondary penalty factor a and Lagrangian r is introduced into variational methods model, it is bright obtains augmentation glug
Day equation；
Wherein, a is secondary penalty factor, and it is default value 2000, r Lagrangian that secondary penalty factor a, which is arranged,；
Step 204, u is initializedk、wkWith r numerical value, is updated using the continuous iteration of alternating direction multipliers method, look for iteration optimization
The saddle point of augmentation Lagrange, determines judgement precision e in sequence1, according to formula is determined, for meeting the feelings of the judgement formula
Condition stops iteration and exports K modal components,
Determine formula specifically:
Wherein, e is taken1=1 × 10-6；
Before step 201, step 2 further include:
Step a is carried out VMD decomposition according to the sensing data of known fault, is passed with the acceleration under the malfunction of outer ring
Sensor data instance, as shown in the table by the centre frequency situation of change of component mode function under observation different K values:
As can be seen from the above table, since the 4th modal components, centre frequency tends to definite value, shows to go out after sample
Show and has crossed decomposing phenomenon, therefore, outer ring fault sample K value takes 4, remaining state sample K value can be similarly obtained, as shown in the table,
Step 3, sensitive factor is assessed:
The sensitive factor for calculating each modal components, assessing each modal components includes fault signature degree；
In the modal components that VMD is analyzed, the existing active constituent comprising fault signature also has the interference such as noise point
Amount.Modal components are assessed using sensitive factor, extract active constituent；
Calculate the sensitive factor λ of each modal components obtained in step 2k,
Wherein, δk=βk- αk, wherein αkFor the related coefficient of modal components and fault-signal, βkFor modal components and normally
The related coefficient of signal, the vibration signal of main shaft, institute when fault-signal is outer ring failure, inner ring failure, rolling element malfunction
State the vibration signal of main shaft when normal signal is normal condition；
It is as shown in the table to calculate each modal components sensitive factor amplitude,
Step 4, signal reconstruction:
Cancelling noise interference, reconstruction signal calculate the feature vector of each reconstruction signal, construct model analysis vector；
It specifically includes:
Step 401, using sort method to sensitive factor λkIt is descending to be ranked up, obtain new sensitive factor sequence
λ′k；
Sort method is bubbling method or cocktail ranking method or radix sorting or Shell sorting, and the present embodiment uses
Bubbling method arranges sensitive factor sequence descending, seeks sensitive factor difference spectrum, as shown in Fig. 2, difference spectrum is at second in figure
Occurs peak value at sensitive factor, therefore selection descending arranges the corresponding modal components of the first two sensitive factor and is overlapped, reconstruct letter
Number；
Step 402, according to sensitive factor sequence λ 'k, calculating difference spectrum assessment vector sequence dk,
dk=λ 'k+1-λ′k；
Step 403, difference spectrum assessment vector sequence d is found outkMaximum value dj；
Step 404, pass through preceding j dkCorresponding sensitive factor λ 'k, corresponding j modal components are found out, j mode is superimposed
Component obtains reconstruction signal；
Step 405, according to reconstruction signal, feature vector is obtained using singular value decomposition method；
Reconstruction signal is divided into 4 rows, feature vector is calculated and constitutes model analysis vector, as shown in figure 3, different main shaft states
Sample of signal model analysis vector trend, it is seen that the feature vector reflection feature distribution of different main shaft states is different.
Step 5, VPMCD fault diagnosis:
Fault identification is carried out to the model analysis vector with prediction model, calculates the error sum of squares of main shaft state VPM,
It is to determine to carry out Classification and Identification with minimum value；
Wherein, VPMCD is the pattern-recognition of variable prediction model, and the main shaft state includes normal, outer ring failure, inner ring
Failure, rolling element failure；
It specifically includes:
Step 501, feature vector is substituted into variable prediction model, variable prediction model is VPMi:
Xi=f (Xj,b0,bj,bjj,bjk)+e2,
Wherein, b0,bj,bjj,bjkFor model parameter；e2To predict error；
Model parameter is obtained using the training of each failure classes vibration data, and prediction error is to preset.Prediction model is
Linear model or linear reciprocal model or secondary model or secondary interaction models；
Linear model are as follows:
Linear reciprocal model are as follows:
Secondary model are as follows:
Secondary interaction models are as follows:
The linear reciprocal model construction VPMCD identification model that the present embodiment uses, by the mode of the training sample of each state
It analyzes vector and substitutes into linear reciprocal model, estimate b0, bj, bjj, bjkIt is corresponding to obtain each feature vector for totally 7 parameter values
SubmodelConstruct the VPMCD identification model of each state, wherein i indicates that Status Type, i=1 represent normal condition
VPM model, 2 represent outer ring failure, and 3 represent inner ring failure, and 4 represent rolling element failure；K representation eigenvalue, k=1 indicate certain shape
The VPM model of first characteristic value of state, and so on.
Step 502, the error sum of squares of each state VPM is calculated, is to determine to carry out Classification and Identification with minimum value；
The model analysis vector for calculating 20 groups of test samples of each state is identified using VPMCD identification model, as a result such as following table
It is shown:
The results show that in addition to 4 groups of outer ring fault samples, remaining is all correctly validated, and accumulative discrimination reaches in test sample
To 98.75%.
The invention has the following beneficial effects: by fan vibration signal carry out VMD analysis, by signal decomposition at one group
The modal components of different frequency, since fault signature generally only occurs at specific frequency range, using sensitive factor to mode
Component carries out analysis and assessment, and screening includes the component of fault signature, excludes noise jamming component, and then reconstruction signal, highlights event
Hinder information；Then the characteristic value constitutive characteristic for extracting reconstruction signal identifies vector, in conjunction with VPMCD method by establishing in characteristic value
Fault diagnosis is carried out in relationship, the selection of traditional classifier subjective parameters and searching process is avoided, shortens failure diagnosis time.
Those skilled in the art is not under conditions of departing from the spirit and scope of the present invention that claims determine, also
Various modifications can be carried out to the above content.Therefore the scope of the present invention is not limited in above explanation, but by
The range of claims determines.
Claims (10)
1. a kind of diagnostic method of wind generator set main shaft failure characterized by comprising
Step 1, vibration signals collecting:
Vibration signal is obtained by the sensor being mounted on wind generator set main shaft seat；
Step 2, VMD variation mode decomposition:
VMD variation mode decomposition is carried out to the vibration signal, obtains the modal components under multiple and different frequencies；
Step 3, sensitive factor is assessed:
The sensitive factor for calculating each modal components, assessing each modal components includes fault signature degree；
Step 4, signal reconstruction:
Cancelling noise interference, reconstruction signal calculate the feature vector of each reconstruction signal, construct model analysis vector；
Step 5, VPMCD fault diagnosis:
Fault identification is carried out to the model analysis vector with prediction model, the error sum of squares of main shaft state VPM is calculated, with most
Small value carries out Classification and Identification for judgement；
Wherein, VMD is variation mode decomposition, and VPMCD is the pattern-recognition of variable prediction model, and the main shaft state includes just
Often, outer ring failure, inner ring failure, rolling element failure.
2. diagnostic method according to claim 1, which is characterized in that sensor described in step 1 is acceleration transducer.
3. diagnostic method according to claim 1, which is characterized in that step 2 specifically includes:
Step 201, the vibration signal is decomposed as K simple component mode function, the simple component mode function is limited in band
Report mode function:
Wherein, f (t) is vibration signal, ukIt (t) is k-th of limit band Intrinsic mode function, AkIt (t) is envelope,For phase
Bit function, t are the time；
Step 202, the frequency bandwidth for estimating each simple component mode function, establishes variational methods model:
Wherein, wkFor k-th of simple component mode function uk(t) center angular frequency, { uk}={ u1……uK, { wk}={ w1……
wK,For the adduction of all simple component mode functions；
Step 203, secondary penalty factor a and Lagrangian r is introduced into variational methods model, obtains augmentation Lagrange's equation；
Wherein, a is secondary penalty factor, r Lagrangian；
Step 204, u is initializedk、wkWith r numerical value, is updated using the continuous iteration of alternating direction multipliers method, look for iteration optimization sequence
The saddle point of middle augmentation Lagrange, determines judgement precision e1, according to formula is determined, the case where for meeting the judgement formula, stop
Only iteration exports K modal components.
4. diagnostic method according to claim 3, which is characterized in that the judgement formula specifically:
Wherein, e is taken1=1 × 10-6。
5. diagnostic method according to claim 1, which is characterized in that step 3 specifically:
Calculate the sensitive factor λ of each modal components obtained in step 2k,
Wherein, δk=βk- αk, wherein αkFor the related coefficient of modal components and fault-signal, βkFor modal components and normal signal
Related coefficient.
6. diagnostic method according to claim 1 or 5, which is characterized in that step 4 specifically includes:
Step 401, using sort method to the sensitive factor λkIt is descending to be ranked up, obtain new sensitive factor sequence λ 'k；
Step 402, according to sensitive factor sequence λ 'k, calculating difference spectrum assessment vector sequence dk,
dk=λ 'k+1-λ′k；
Step 403, the difference spectrum assessment vector sequence d is found outkMaximum value dj；
Step 404, pass through preceding j dkCorresponding sensitive factor λ 'k, corresponding j modal components are found out, the j mode is superimposed
Component obtains reconstruction signal；
Step 405, according to reconstruction signal, feature vector is obtained using singular value decomposition method.
7. diagnostic method according to claim 6, which is characterized in that sort method described in step 401 be bubbling method or
Person's cocktail ranking method or radix sorting or Shell sorting.
8. diagnostic method according to claim 1, which is characterized in that step 5 specifically includes:
Step 501, feature vector is substituted into variable prediction model, the variable prediction model is VPMi:
Xi=f (Xj,b0,bj,bjj,bjk)+e2,
Wherein, b0,bj,bjj,bjkFor model parameter；e2To predict error；
Step 502, the error sum of squares of each state VPM is calculated, is to determine to carry out Classification and Identification with minimum value.
9. diagnostic method according to claim 8, which is characterized in that the model parameter is to utilize each failure classes vibration number
It is obtained according to training, the prediction error is to preset.
10. diagnostic method according to claim 8, which is characterized in that the prediction model is linear model or linear friendship
Mutual model or secondary model or secondary interaction models；
The linear model are as follows:
The linear reciprocal model are as follows:
The secondary model are as follows:
The secondary interaction models are as follows:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710423126.3A CN109000921B (en) | 2017-06-07 | 2017-06-07 | Method for diagnosing main shaft fault of wind turbine generator |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710423126.3A CN109000921B (en) | 2017-06-07 | 2017-06-07 | Method for diagnosing main shaft fault of wind turbine generator |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109000921A true CN109000921A (en) | 2018-12-14 |
| CN109000921B CN109000921B (en) | 2020-05-08 |
Family
ID=64573004
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710423126.3A Active CN109000921B (en) | 2017-06-07 | 2017-06-07 | Method for diagnosing main shaft fault of wind turbine generator |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109000921B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109489977A (en) * | 2018-12-28 | 2019-03-19 | 西安工程大学 | Method for Bearing Fault Diagnosis based on KNN-AdaBoost |
| CN110186682A (en) * | 2019-07-08 | 2019-08-30 | 石家庄铁道大学 | Fault Diagnosis of Roller Bearings based on fractional order variation mode decomposition |
| CN111413089A (en) * | 2020-04-08 | 2020-07-14 | 北华大学 | Gear fault diagnosis method based on combination of VMD entropy method and VPMCD |
| CN111985315A (en) * | 2020-07-10 | 2020-11-24 | 合肥工业大学 | Bearing fault signal intrinsic mode function decomposition and extraction method and device |
| CN112326245A (en) * | 2020-10-21 | 2021-02-05 | 中国航空工业集团公司上海航空测控技术研究所 | Rolling bearing fault diagnosis method based on variational Hilbert-Huang transform |
| CN116625688A (en) * | 2023-05-24 | 2023-08-22 | 石家庄铁道大学 | Rolling bearing health monitoring method based on multilayer noise reduction and self-encoder |
| CN116738372A (en) * | 2023-08-15 | 2023-09-12 | 昆仑数智科技有限责任公司 | Rolling bearing fault diagnosis method, device and equipment for refining centrifugal pump |
| CN117030268A (en) * | 2023-10-07 | 2023-11-10 | 太原科技大学 | Rolling bearing fault diagnosis method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105758644A (en) * | 2016-05-16 | 2016-07-13 | 上海电力学院 | Rolling bearing fault diagnosis method based on variation mode decomposition and permutation entropy |
| CN106017926A (en) * | 2016-05-13 | 2016-10-12 | 山东理工大学 | Rolling bearing fault diagnosis method based on variational mode decomposition |
| CN106198015A (en) * | 2016-06-29 | 2016-12-07 | 潍坊学院 | The VMD of a kind of rolling bearing, spectrum kurtosis and smooth iteration envelope Analysis Method |
| EP2810027B1 (en) * | 2012-01-30 | 2017-06-07 | S.P.M. Instrument AB | Apparatus and method for analysing the condition of a machine having a rotating part |
-
2017
- 2017-06-07 CN CN201710423126.3A patent/CN109000921B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2810027B1 (en) * | 2012-01-30 | 2017-06-07 | S.P.M. Instrument AB | Apparatus and method for analysing the condition of a machine having a rotating part |
| CN106017926A (en) * | 2016-05-13 | 2016-10-12 | 山东理工大学 | Rolling bearing fault diagnosis method based on variational mode decomposition |
| CN105758644A (en) * | 2016-05-16 | 2016-07-13 | 上海电力学院 | Rolling bearing fault diagnosis method based on variation mode decomposition and permutation entropy |
| CN106198015A (en) * | 2016-06-29 | 2016-12-07 | 潍坊学院 | The VMD of a kind of rolling bearing, spectrum kurtosis and smooth iteration envelope Analysis Method |
Non-Patent Citations (2)
| Title |
|---|
| 程军圣等: "基于VPMCD 和EMD 的齿轮故障诊断方法", 《振动与冲击》 * |
| 郑小霞等: "基于变分模态分解的风机滚动轴承早期故障诊断", 《轴承》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109489977A (en) * | 2018-12-28 | 2019-03-19 | 西安工程大学 | Method for Bearing Fault Diagnosis based on KNN-AdaBoost |
| CN110186682A (en) * | 2019-07-08 | 2019-08-30 | 石家庄铁道大学 | Fault Diagnosis of Roller Bearings based on fractional order variation mode decomposition |
| CN111413089A (en) * | 2020-04-08 | 2020-07-14 | 北华大学 | Gear fault diagnosis method based on combination of VMD entropy method and VPMCD |
| CN111985315A (en) * | 2020-07-10 | 2020-11-24 | 合肥工业大学 | Bearing fault signal intrinsic mode function decomposition and extraction method and device |
| CN112326245A (en) * | 2020-10-21 | 2021-02-05 | 中国航空工业集团公司上海航空测控技术研究所 | Rolling bearing fault diagnosis method based on variational Hilbert-Huang transform |
| CN112326245B (en) * | 2020-10-21 | 2023-03-10 | 中国航空工业集团公司上海航空测控技术研究所 | Rolling bearing fault diagnosis method based on variational Hilbert-Huang transform |
| CN116625688A (en) * | 2023-05-24 | 2023-08-22 | 石家庄铁道大学 | Rolling bearing health monitoring method based on multilayer noise reduction and self-encoder |
| CN116738372A (en) * | 2023-08-15 | 2023-09-12 | 昆仑数智科技有限责任公司 | Rolling bearing fault diagnosis method, device and equipment for refining centrifugal pump |
| CN116738372B (en) * | 2023-08-15 | 2023-10-27 | 昆仑数智科技有限责任公司 | Rolling bearing fault diagnosis method, device and equipment for refining centrifugal pump |
| CN117030268A (en) * | 2023-10-07 | 2023-11-10 | 太原科技大学 | Rolling bearing fault diagnosis method |
| CN117030268B (en) * | 2023-10-07 | 2024-01-23 | 太原科技大学 | Rolling bearing fault diagnosis method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109000921B (en) | 2020-05-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109000921A (en) | A kind of diagnostic method of wind generator set main shaft failure | |
| CN105760839A (en) | Bearing fault diagnosis method based on multi-feature manifold learning and support vector machine | |
| CN106197999B (en) | A kind of planetary gear method for diagnosing faults | |
| Hu et al. | A novel fault diagnosis technique for wind turbine gearbox | |
| CN100485342C (en) | Integrated supporting vector machine mixed intelligent diagnosing method for mechanical fault | |
| CN103115789B (en) | Second generation small-wave support vector machine assessment method for damage and remaining life of metal structure | |
| WO2023123593A1 (en) | Variational mode decomposition and residual network-based aviation bearing fault diagnosis method | |
| CN104614179B (en) | A kind of gearbox of wind turbine state monitoring method | |
| CN107016404A (en) | Wind power generating set failure prediction method based on D S evidence fusions | |
| CN109827777A (en) | Rolling bearing fault prediction technique based on Partial Least Squares extreme learning machine | |
| CN107103658B (en) | Aerial engine air passage method for detecting abnormality based on deep learning and Gaussian Profile | |
| CN106021771A (en) | Method and device for diagnosing faults | |
| CN106644484A (en) | Turboprop Engine rotor system fault diagnosis method through combination of EEMD and neighborhood rough set | |
| CN104655425A (en) | Bearing fault classification diagnosis method based on sparse representation and LDM (large margin distribution machine) | |
| CN109255333B (en) | Fault hybrid diagnosis method for rolling bearing of large wind turbine generator | |
| Ramirez et al. | Alarms management by supervisory control and data acquisition system for wind turbines | |
| CN110146289A (en) | A kind of rolling bearing Weak fault feature extracting method | |
| CN107101829A (en) | A kind of intelligent diagnosing method of aero-engine structure class failure | |
| Shang et al. | Fault diagnosis method of rolling bearing based on deep belief network | |
| Wang et al. | A quantitative intelligent diagnosis method for early weak faults of aviation high-speed bearings | |
| CN113357098A (en) | Fault early warning method for fan variable pitch subsystem | |
| Wang et al. | Novel paralleled extreme learning machine networks for fault diagnosis of wind turbine drivetrain | |
| CN109325553B (en) | Wind power gear box fault detection method, system, equipment and medium | |
| Shen et al. | Fault detection method based on multi-scale convolutional neural network for wind turbine gearbox | |
| Ren et al. | A clustering-based method for health conditions evaluation of aero-engines |
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 |