CN106777606A - A kind of gearbox of wind turbine failure predication diagnosis algorithm - Google Patents

A kind of gearbox of wind turbine failure predication diagnosis algorithm Download PDF

Info

Publication number
CN106777606A
CN106777606A CN201611097387.2A CN201611097387A CN106777606A CN 106777606 A CN106777606 A CN 106777606A CN 201611097387 A CN201611097387 A CN 201611097387A CN 106777606 A CN106777606 A CN 106777606A
Authority
CN
China
Prior art keywords
data
gearbox
wind turbine
model
sample
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
Application number
CN201611097387.2A
Other languages
Chinese (zh)
Inventor
丁云飞
王栋璀
朱晨烜
刘洋
潘羿龙
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Dianji University
Original Assignee
Shanghai Dianji University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shanghai Dianji University filed Critical Shanghai Dianji University
Priority to CN201611097387.2A priority Critical patent/CN106777606A/en
Publication of CN106777606A publication Critical patent/CN106777606A/en
Pending legal-status Critical Current

Links

Classifications

    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
    • G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Geometry (AREA)
  • Computer Hardware Design (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses the failure predication diagnostic method the invention provides gearbox of wind turbine of the one kind based on core principle component analysis (KPCA) and SVMs, before and after when having taken into full account that gearbox fault occurs, each component temperature change and failure occur on the basis of the index such as power output change, input dimension is reduced with KPCA algorithms carries out feature extraction, give up incoherent data, model training speed can be greatly improved, failure diagnosis time is reduced.Introducing supporting vector function simultaneously carries out classification based training raising generalization ability to data, and result is analyzed and explained by expert system, accurate and full and accurate information can be provided for human-computer interaction interface, it is achieved thereby that to the Precise Diagnosis of failure.

Description

A kind of gearbox of wind turbine failure predication diagnosis algorithm
Technical field
The present invention relates to fault diagnosis algorithm field, specifically, it is related specifically to a kind of based on KPCA and supporting vector The gearbox of wind turbine failure predication diagnosis algorithm of machine.
Background technology
In the past few years, country attaches great importance to generation of electricity by new energy.Wind-powered electricity generation is with a kind of generation of electricity by new energy skill of clean environment firendly the most Art and by extensive concern.The fast development of wind-powered electricity generation also brings unprecedented problem-fault diagnosis to wind-powered electricity generation equipment manufacturing. And the drive gearbox in mechanical system is one of fault rate highest equipment in Wind turbines, its running status will be straight Connect the running status and power output of influence Wind turbines, it is therefore necessary to which the fault diagnosis to gearbox of wind turbine carries out depth Enter research.
At present, the method for diagnosing faults of proposition mainly includes two classes.One is the vibration according to gear-box under operation Signal to the fault type grasped carries out pattern-recognition；Two is that possible generation is predicted by its dependent variable in gear-box Failure, and diagnosis is made to failure by neutral net.But for the high dimensional data of multivariable, neutral net easily falls into Enter local minimum problem, it may appear that cross adaptation.SVMs (SVM) then to the high dimensional data processing speed of multivariable compared with Slowly, operation time is more long, causes diagnosis to slow.It is therefore desirable to propose that one kind can improve training sample generalization ability By force, while taking into account the new method of diagnosis speed and diagnostic accuracy.
The content of the invention
It is an object of the invention to be directed to deficiency of the prior art, there is provided a kind of gearbox of wind turbine failure predication is examined Disconnected algorithm, power output change before and after each component temperature change and failure occur when having taken into full account that gearbox fault occurs On the basis of etc. index, input dimension is reduced with KPCA algorithms carries out feature extraction, gives up incoherent data, can be significantly Model training speed is improved, failure diagnosis time is reduced.Carried while introducing supporting vector function and carrying out classification based training to data Generalization ability high, and result is analyzed and is explained by expert system, can for human-computer interaction interface provide it is accurate and And full and accurate information, it is achieved thereby that to the Precise Diagnosis of failure.
Technical problem solved by the invention can be realized using following technical scheme：
A kind of gearbox of wind turbine failure predication diagnosis algorithm, comprises the following steps：
1) temperature, wind that operating gear-box oil temperature, cabin temperature, gear-box drive front end and non-driven front end are obtained Speed, the history samples time data of blower fan power output；
2) history samples time data is pre-processed, including removal data exception point and data normalization；
3) dimensionality reduction is carried out to pretreated history samples time data using KPCA algorithms, the feature input that will be extracted As the training sample set and test sample collection of model；
4) training sample is modeled using SVMs；
5) parameter of SVMs is optimized using cross-validation method, obtains optimal forecast model, circulation changes For training pattern；
6) result of prediction is sent in expert system to be analyzed explanation and draw a conclusion and is presented on human-computer interaction interface On.
The step 3) including following flow：
3.1) raw sample data of the n index that will be obtained constitutes the matrix of one (m × n) dimension, and wherein m is each The sample number of index.
3.2) nuclear matrix is calculated, the parameter in Gauss radial direction kernel function is first selected, nuclear matrix K is calculated；
3.3) nuclear matrix is corrected, by nuclear matrix centralization；
3.4) eigenvalue λ of nuclear matrix is calculated with the method for Jacobi iteration1..., λn, corresponding characteristic vector is V1..., Vn；
3.5) by Eigenvector normalization, characteristic value is sorted to obtain λ ' in descending order1＞ ... ＞ λ 'n, and adjust characteristic vector and enter Capable V '1..., V 'n, unit orthogonalized eigenvectors obtain a1..., an；
3.6) the accumulation contribution rate B of characteristic value is calculated1..., Bn, according to extraction efficiency P set in advance, work as Bt>=P, then T principal component a before extracting1..., at；
3.7) corresponding larger characteristic value and characteristic vector are selected, sample matrix is then calculated in higher dimensional space at these Projection in characteristic vector, the projection of gained is exactly sample data the data obtained after core principle component KPCA dimensionality reductions.
The step 4) including following flow：
4.1) supporting vector machine model for setting up training sample is represented by：
Wherein, ω is model parameter i.e. weight vectors,It is that nonlinear characteristic from the input space to higher dimensional space is reflected Penetrate, b is residual error.
4.2) Lagrange multiplier λ is introducediAfterwards, the forecast model based on SVM is changed into：
Wherein, kernel function K () uses gaussian radial basis function form：
K (x, xi)=exp (- | | x-xi||2/σ2)
Wherein, xi (i=1,2,3 ..., N) is input training sample, and σ is kernel functional parameter.
The step 5) also include：
5.1) feature input set is divided into K each group, as K subset；
5.2) each subset data is made into one-time authentication collection respectively, remaining K-1 groups subset can so be obtained as training set To K model；
5.3) average of the classification accuracy collected with the final checking of K model as the support under cross-validation method to The performance indications of amount machine.
The step 6) it is further comprising the steps of：
6.1) will predict the outcome and be sent to fault message storehouse and preserve and export to inference machine；
6.2) output of the inference machine to fault message storehouse makes inferences analysis, and the rule in knowledge base is matched repeatedly, so that Obtain corresponding failure cause；
6.3) failure cause is sent in interpreter, route and conclusion by inference are given to be explained and present accordingly On human-computer interaction interface, user is set to can be clearly seen that reasoning process.
Compared with prior art, beneficial effects of the present invention are as follows：
The present invention takes full advantage of extractabilities of the KPCA to nonlinear characteristic vector, gives up uncorrelated data, not only subtracts Lacked the parameter of SVMs needs prediction, and be effectively improved the training time, optimize forecast model speed and Precision.Additionally, SVMs can process the problem of nonlinear regression, with globally optimal solution, neutral net is solved Local minimum problem and excessively adaptation, can well improve the generalization ability and precision of prediction of model.By means of simultaneously specially Family's system so that predict the outcome and be obtained for reasoning and explanation, solve the troubleshooting issue of Wind turbines gear mechanism, be wind The safe and reliable operation of group of motors is provided and ensured, such that it is able to optimize dispatching of power netwoks, realizes the safe and stable and economy of power network Operation.
Brief description of the drawings
Fig. 1 is the flow chart of gearbox of wind turbine failure predication diagnostic model of the present invention.
Fig. 2 is expert system module structure chart of the present invention.
Specific embodiment
For technological means, creation characteristic, reached purpose and effect for making present invention realization are easy to understand, with reference to Specific embodiment, is expanded on further the present invention.
Fig. 1 is fault diagnosis model flow chart of the invention.Operating historical data is obtained first, and then data are entered Row pretreatment, secondly carries out feature extraction to multivariable to reduce dimension using core principle component analysis (KPCA), again with extraction Feature input out sets up SVMs (SVM) as training sample, and parameter is optimized by cross-validation method To improve the generalization ability of model, finally the result of prediction is sent to be analyzed explanation and draw a conclusion in expert system and is in On present human-computer interaction interface.Its specific embodiment comprises the following steps：
Step 1, obtains operating gear-box oil temperature, and cabin temperature, gear-box drives the temperature of front end and non-driven front end Degree, bearing temperature, wind speed round, the history samples time data of blower fan power output.
Data are pre-processed by step 2.
Preferably, should also be pre-processed to data using normalization formula in step 2.Normalization formula is as follows：
xmax-xmin
Data are carried out dimensionality reduction by step 3 using KPCA algorithms, training of the characteristic vector input that will be extracted as model Sample set and test sample collection.
Further, it is further comprising the steps of in step 3：
Step 3.1, the raw sample data of the n index that will be obtained constitutes the matrix of one (m × n) dimension, and wherein m is The sample number of each index.
Step 3.2, calculates nuclear matrix, first selectes the parameter in Gauss radial direction kernel function, calculates nuclear matrix K.
Step 3.3, corrects nuclear matrix, by nuclear matrix centralization.
Step 3.4, the eigenvalue λ of nuclear matrix is calculated with the method for Jacobi iteration1..., λn, corresponding characteristic vector It is V1..., Vn。
Step 3.5, Eigenvector normalization sorts to obtain λ ' characteristic value in descending order1＞ ... ＞ λ 'n, and adjust feature to Amount carries out obtaining V '1..., V 'n, unit orthogonalized eigenvectors obtain a1..., an。
Step 3.6, calculates the accumulation contribution rate B of characteristic value1..., Bn, according to extraction efficiency P set in advance, work as Bt≥ P, then t principal component a before extracting1..., at。
Step 3.7, selects corresponding larger characteristic value and characteristic vector, then calculate sample matrix in higher dimensional space Projection in these characteristic vectors.The projection of gained is exactly sample data the data obtained after core principle component KPCA dimensionality reductions.
Step 4, SVMs (SVM) is set up using the feature input for extracting, and by cross-validation method to parameter Optimize and model is trained with training sample.
Further, step 4 should also include：
Step 4.1, the supporting vector machine model for setting up training sample is represented by：
Wherein, ω is model parameter i.e. weight vectors,It is that nonlinear characteristic from the input space to higher dimensional space is reflected Penetrate, b is residual error.
Step 4.2, introduces Lagrange multiplier λiAfterwards, the forecast model based on SVM is changed into：
Its Kernel Function K () uses gaussian radial basis function form：
K (x, xi)=exp (- | | x-xi||2/σ2)
Wherein xi (i=1,2,3 ..., N) is input training sample, and σ is kernel functional parameter.
Step 5, is optimized using cross-validation method to the parameter of SVMs, obtains optimal forecast model, is followed Ring iterative training pattern.
Further, the performance parameter optimal to determine SVMs, step 5 is further comprising the steps of：
Feature input set is divided into K each group by step 5.1, as K subset.
Each subset data is made one-time authentication collection by step 5.2 respectively, remaining K-1 groups subset as training set, so K model can be obtained.
Step 5.3 average of the classification accuracy of the final checking collection of this K model is used as under cross-validation method The performance indications of SVMs.
Step 6, the result of prediction is sent in expert system to be analyzed explanation and draw a conclusion is presented on man-machine friendship On mutual interface.
Further by Fig. 2 expert system flow charts, step 6 is further comprising the steps of：
Step 6.1 will predict the outcome and be sent to fault message storehouse and preserve and export to inference machine.
Output of step 6.2 inference machine to fault message storehouse makes inferences analysis, and the rule in knowledge base is matched repeatedly, from And obtain corresponding failure cause.
Be sent to failure cause in interpreter by step 6.3, and route and conclusion by inference provide corresponding explanation simultaneously It is presented on human-computer interaction interface, user is can be clearly seen that reasoning process.
General principle of the invention and principal character and advantages of the present invention has been shown and described above.The technology of the industry Personnel it should be appreciated that the present invention is not limited to the above embodiments, simply explanation described in above-described embodiment and specification this The principle of invention, without departing from the spirit and scope of the present invention, various changes and modifications of the present invention are possible, these changes Change and improvement all fall within the protetion scope of the claimed invention.The claimed scope of the invention by appending claims and its Equivalent thereof.

Claims (5)

1. a kind of gearbox of wind turbine failure predication diagnosis algorithm, it is characterised in that comprise the following steps：
1) obtain operating gear-box oil temperature, cabin temperature, gear-box drive the temperature of front end and non-driven front end, wind speed, The history samples time data of blower fan power output；
2) history samples time data is pre-processed, including removal data exception point and data normalization；
3) dimensionality reduction is carried out to pretreated history samples time data using KPCA algorithms, the input of the feature that will extract as The training sample set and test sample collection of model；
4) training sample is modeled using SVMs；
5) parameter of SVMs is optimized using cross-validation method, obtains optimal forecast model, loop iteration instruction Practice model；
6) result of prediction is sent in expert system to be analyzed explanation and draw a conclusion and is presented on human-computer interaction interface.
2. gearbox of wind turbine failure predication diagnosis algorithm according to claim 1, it is characterised in that the step 3) Including following flow：
3.1) raw sample data of the n index that will be obtained constitutes the matrix of one (m × n) dimension, and wherein m is each index Sample number.
3.2) nuclear matrix is calculated, the parameter in Gauss radial direction kernel function is first selected, nuclear matrix K is calculated；
3.3) nuclear matrix is corrected, by nuclear matrix centralization；
3.4) eigenvalue λ of nuclear matrix is calculated with the method for Jacobi iteration1..., λn, corresponding characteristic vector is V1..., Vn；
3.5) by Eigenvector normalization, characteristic value is sorted to obtain λ ' in descending order1＞ ... ＞ λ 'n, and adjust characteristic vector and carry out V′1..., V 'n, unit orthogonalized eigenvectors obtain a1..., an；
3.6) the accumulation contribution rate B of characteristic value is calculated1..., Bn, according to extraction efficiency P set in advance, work as Bt>=P, then extract Preceding t principal component a1..., at；
3.7) corresponding larger characteristic value and characteristic vector are selected, sample matrix is then calculated in higher dimensional space in these features Projection on vector, the projection of gained is exactly sample data the data obtained after core principle component KPCA dimensionality reductions.
3. gearbox of wind turbine failure predication diagnosis algorithm according to claim 1, it is characterised in that the step 4) Including following flow：
4.1) supporting vector machine model for setting up training sample is represented by：
Wherein, ω is model parameter i.e. weight vectors,It is the nonlinear characteristic mapping from the input space to higher dimensional space, b is Residual error.
4.2) Lagrange multiplier λ is introducediAfterwards, the forecast model based on SVM is changed into：
f ( x ) = Σ i = 1 N λ i K ( x , x i ) + b
Wherein, kernel function K () uses gaussian radial basis function form：
K (x, xi)=exp (- | | x-xi||2/σ2)
Wherein, xi (i=1,2,3 ..., N) is input training sample, and σ is kernel functional parameter.
4. gearbox of wind turbine failure predication diagnosis algorithm according to claim 1, it is characterised in that the step 5) Also include：
5.1) feature input set is divided into K each group, as K subset；
5.2) each subset data is made into one-time authentication collection respectively, remaining K-1 groups subset can so obtain K as training set Individual model；
5.3) average of the classification accuracy collected with the K final checking of model is used as the SVMs under cross-validation method Performance indications.
5. gearbox of wind turbine failure predication diagnosis algorithm according to claim 1, it is characterised in that the step 6) It is further comprising the steps of：
6.1) will predict the outcome and be sent to fault message storehouse and preserve and export to inference machine；
6.2) output of the inference machine to fault message storehouse makes inferences analysis, the rule in knowledge base is matched repeatedly, so as to obtain Corresponding failure cause；
6.3) failure cause is sent in interpreter, route and conclusion by inference are given to be explained and be presented on people accordingly On machine interactive interface, user is set to can be clearly seen that reasoning process.
CN201611097387.2A 2016-12-02 2016-12-02 A kind of gearbox of wind turbine failure predication diagnosis algorithm Pending CN106777606A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201611097387.2A CN106777606A (en) 2016-12-02 2016-12-02 A kind of gearbox of wind turbine failure predication diagnosis algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201611097387.2A CN106777606A (en) 2016-12-02 2016-12-02 A kind of gearbox of wind turbine failure predication diagnosis algorithm

Publications (1)

Publication Number Publication Date
CN106777606A true CN106777606A (en) 2017-05-31

Family

ID=58883183

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201611097387.2A Pending CN106777606A (en) 2016-12-02 2016-12-02 A kind of gearbox of wind turbine failure predication diagnosis algorithm

Country Status (1)

Country Link
CN (1) CN106777606A (en)

Cited By (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107392248A (en) * 2017-07-21 2017-11-24 重庆青山工业有限责任公司 Gear parameter Contribution Analysis method based on PCA reconstruction errors
CN108009691A (en) * 2017-12-22 2018-05-08 东软集团股份有限公司 Equipment life Forecasting Methodology, device and equipment
CN108073158A (en) * 2017-12-05 2018-05-25 上海电机学院 Based on PCA and KNN density algorithm Wind turbines Method for Bearing Fault Diagnosis
CN108182302A (en) * 2017-12-13 2018-06-19 东北大学 Incipient fault detection method based on modification cluster semi-supervised kernel pivot analysis
CN108223782A (en) * 2018-01-15 2018-06-29 山东科技大学 A kind of epicyclic gearbox fault detection method
CN108509732A (en) * 2018-04-03 2018-09-07 厦门理工学院 Appraisal procedure, terminal device and the storage medium of steam turbine fault severity level
CN108805195A (en) * 2018-04-03 2018-11-13 北京理工大学 A kind of motor group method for diagnosing faults based on two-value deep-neural-network
CN108875558A (en) * 2018-04-27 2018-11-23 浙江师范大学 A kind of high-performance large-scale wind turbine gearbox Fault Classification and system
CN108898182A (en) * 2018-07-02 2018-11-27 武汉科技大学 A kind of MMC method for diagnosing faults based on core pivot element analysis and support vector machines
CN108921230A (en) * 2018-07-25 2018-11-30 浙江浙能嘉华发电有限公司 Method for diagnosing faults based on class mean value core pivot element analysis and BP neural network
CN109238691A (en) * 2017-07-06 2019-01-18 神华集团有限责任公司 Portable fault detection system, fault detection method
CN109711053A (en) * 2018-12-27 2019-05-03 北京航天智造科技发展有限公司 Based on industrial big data Research on Fan Fault Forecasting method and system
CN109738776A (en) * 2019-01-02 2019-05-10 华南理工大学 Fan converter open-circuit fault recognition methods based on LSTM
CN110082136A (en) * 2019-04-19 2019-08-02 电子科技大学 Rotary machinery fault diagnosis method based on Retrieval method Support Vector Machines Optimized
CN110442833A (en) * 2019-06-10 2019-11-12 内蒙古工业大学 One kind assessing Wind turbines health state evaluation method based on various dimensions SCADA data
CN110735769A (en) * 2019-09-18 2020-01-31 西安察柏科技咨询有限公司 method, device and system for predicting fan faults
CN110940514A (en) * 2018-09-20 2020-03-31 中车株洲电力机车研究所有限公司 Method for constructing and diagnosing early diagnosis model of fan gear box heat dissipation fault
CN111539482A (en) * 2020-04-28 2020-08-14 三峡大学 RBF kernel function-based space multidimensional wind power data dimension reduction and reconstruction method
CN111581597A (en) * 2020-03-17 2020-08-25 华电电力科学研究院有限公司 Wind turbine generator gearbox bearing temperature state monitoring method based on self-organizing kernel regression model
CN111856209A (en) * 2020-07-23 2020-10-30 广东电网有限责任公司清远供电局 Power transmission line fault classification method and device
WO2020228025A1 (en) * 2019-05-16 2020-11-19 西门子股份公司 Method and apparatus for performing modeling analysis on data of type of machine tools
CN112232405A (en) * 2020-10-13 2021-01-15 中车青岛四方机车车辆股份有限公司 Fault prediction, monitoring and diagnosis method of gearbox and corresponding device
CN113532848A (en) * 2021-06-24 2021-10-22 南京工业大学 Fault diagnosis system for planetary gear box
CN115146718A (en) * 2022-06-27 2022-10-04 北京华能新锐控制技术有限公司 Depth representation-based wind turbine generator anomaly detection method
CN115169405A (en) * 2022-07-14 2022-10-11 北京威控科技股份有限公司 Hotel guest room equipment fault diagnosis method and system based on support vector machine

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105136454A (en) * 2015-10-15 2015-12-09 上海电机学院 Wind turbine gear box fault recognition method
CN105354587A (en) * 2015-09-25 2016-02-24 国网甘肃省电力公司电力科学研究院 Fault diagnosis method for gearbox of wind generation unit
CN105760617A (en) * 2016-03-07 2016-07-13 华北电力大学（保定） Calculation method applied to multi-parameter fault prediction and judgment indexes of wind generating set

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105354587A (en) * 2015-09-25 2016-02-24 国网甘肃省电力公司电力科学研究院 Fault diagnosis method for gearbox of wind generation unit
CN105136454A (en) * 2015-10-15 2015-12-09 上海电机学院 Wind turbine gear box fault recognition method
CN105760617A (en) * 2016-03-07 2016-07-13 华北电力大学（保定） Calculation method applied to multi-parameter fault prediction and judgment indexes of wind generating set

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
杨帆，马洁: "基于KPCA数据重构的参数优化方法", 《北京信息科技大学学报》 *
赵洪山，胡庆春，李志为: "基于统计过程控制的风机齿轮箱故障预测", 《电力系统保护与控制》 *

Cited By (35)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109238691B (en) * 2017-07-06 2020-05-26 国家能源投资集团有限责任公司 Portable fault detection system and fault detection method
CN109238691A (en) * 2017-07-06 2019-01-18 神华集团有限责任公司 Portable fault detection system, fault detection method
CN107392248B (en) * 2017-07-21 2020-04-24 重庆青山工业有限责任公司 Gear parameter contribution degree analysis method based on PCA reconstruction error
CN107392248A (en) * 2017-07-21 2017-11-24 重庆青山工业有限责任公司 Gear parameter Contribution Analysis method based on PCA reconstruction errors
CN108073158A (en) * 2017-12-05 2018-05-25 上海电机学院 Based on PCA and KNN density algorithm Wind turbines Method for Bearing Fault Diagnosis
CN108182302A (en) * 2017-12-13 2018-06-19 东北大学 Incipient fault detection method based on modification cluster semi-supervised kernel pivot analysis
CN108182302B (en) * 2017-12-13 2021-06-22 东北大学 Early fault detection method based on modified clustering semi-supervised kernel principal component analysis
CN108009691A (en) * 2017-12-22 2018-05-08 东软集团股份有限公司 Equipment life Forecasting Methodology, device and equipment
CN108009691B (en) * 2017-12-22 2021-06-25 东软集团股份有限公司 Equipment life prediction method, device and equipment
CN108223782B (en) * 2018-01-15 2019-07-12 山东科技大学 A kind of epicyclic gearbox fault detection method
CN108223782A (en) * 2018-01-15 2018-06-29 山东科技大学 A kind of epicyclic gearbox fault detection method
CN108805195A (en) * 2018-04-03 2018-11-13 北京理工大学 A kind of motor group method for diagnosing faults based on two-value deep-neural-network
CN108509732A (en) * 2018-04-03 2018-09-07 厦门理工学院 Appraisal procedure, terminal device and the storage medium of steam turbine fault severity level
CN108875558A (en) * 2018-04-27 2018-11-23 浙江师范大学 A kind of high-performance large-scale wind turbine gearbox Fault Classification and system
CN108898182A (en) * 2018-07-02 2018-11-27 武汉科技大学 A kind of MMC method for diagnosing faults based on core pivot element analysis and support vector machines
CN108921230A (en) * 2018-07-25 2018-11-30 浙江浙能嘉华发电有限公司 Method for diagnosing faults based on class mean value core pivot element analysis and BP neural network
CN110940514B (en) * 2018-09-20 2021-09-03 中车株洲电力机车研究所有限公司 Method for constructing and diagnosing early diagnosis model of fan gear box heat dissipation fault
CN110940514A (en) * 2018-09-20 2020-03-31 中车株洲电力机车研究所有限公司 Method for constructing and diagnosing early diagnosis model of fan gear box heat dissipation fault
CN109711053A (en) * 2018-12-27 2019-05-03 北京航天智造科技发展有限公司 Based on industrial big data Research on Fan Fault Forecasting method and system
CN109738776A (en) * 2019-01-02 2019-05-10 华南理工大学 Fan converter open-circuit fault recognition methods based on LSTM
CN110082136A (en) * 2019-04-19 2019-08-02 电子科技大学 Rotary machinery fault diagnosis method based on Retrieval method Support Vector Machines Optimized
CN110082136B (en) * 2019-04-19 2021-03-30 电子科技大学 Rotary machine fault diagnosis method based on cloud genetic algorithm optimization support vector machine
WO2020228025A1 (en) * 2019-05-16 2020-11-19 西门子股份公司 Method and apparatus for performing modeling analysis on data of type of machine tools
CN110442833A (en) * 2019-06-10 2019-11-12 内蒙古工业大学 One kind assessing Wind turbines health state evaluation method based on various dimensions SCADA data
CN110442833B (en) * 2019-06-10 2022-09-09 内蒙古工业大学 Wind turbine health state assessment method based on multi-dimensional SCADA data
CN110735769A (en) * 2019-09-18 2020-01-31 西安察柏科技咨询有限公司 method, device and system for predicting fan faults
CN111581597A (en) * 2020-03-17 2020-08-25 华电电力科学研究院有限公司 Wind turbine generator gearbox bearing temperature state monitoring method based on self-organizing kernel regression model
CN111539482A (en) * 2020-04-28 2020-08-14 三峡大学 RBF kernel function-based space multidimensional wind power data dimension reduction and reconstruction method
CN111856209A (en) * 2020-07-23 2020-10-30 广东电网有限责任公司清远供电局 Power transmission line fault classification method and device
CN112232405A (en) * 2020-10-13 2021-01-15 中车青岛四方机车车辆股份有限公司 Fault prediction, monitoring and diagnosis method of gearbox and corresponding device
CN112232405B (en) * 2020-10-13 2022-09-02 中车青岛四方机车车辆股份有限公司 Fault prediction, monitoring and diagnosis method of gearbox and corresponding device
CN113532848A (en) * 2021-06-24 2021-10-22 南京工业大学 Fault diagnosis system for planetary gear box
CN115146718A (en) * 2022-06-27 2022-10-04 北京华能新锐控制技术有限公司 Depth representation-based wind turbine generator anomaly detection method
CN115169405A (en) * 2022-07-14 2022-10-11 北京威控科技股份有限公司 Hotel guest room equipment fault diagnosis method and system based on support vector machine
CN115169405B (en) * 2022-07-14 2024-02-02 北京威控科技股份有限公司 Hotel guest room equipment fault diagnosis method and system based on support vector machine

Similar Documents

Publication Publication Date Title
CN106777606A (en) A kind of gearbox of wind turbine failure predication diagnosis algorithm
Wu et al. Data-driven remaining useful life prediction via multiple sensor signals and deep long short-term memory neural network
Qin et al. Macroscopic–microscopic attention in LSTM networks based on fusion features for gear remaining life prediction
Wang et al. Modified multiscale weighted permutation entropy and optimized support vector machine method for rolling bearing fault diagnosis with complex signals
Zhang et al. Instance-based ensemble deep transfer learning network: A new intelligent degradation recognition method and its application on ball screw
Li et al. Fault diagnosis of rotating machinery based on combination of deep belief network and one-dimensional convolutional neural network
Pandhare et al. Convolutional neural network based rolling-element bearing fault diagnosis for naturally occurring and progressing defects using time-frequency domain features
CN105354587B (en) A kind of method for diagnosing faults of wind-driven generator group wheel box
CN108896296A (en) A kind of wind turbine gearbox method for diagnosing faults based on convolutional neural networks
CN112149316A (en) Aero-engine residual life prediction method based on improved CNN model
Xu et al. Dilated convolution neural network for remaining useful life prediction
CN105628425A (en) Rotation machinery early stage fault diagnosis method based on heredity annealing optimization multi-core support vector machine
WO2021257128A2 (en) Quantum computing based deep learning for detection, diagnosis and other applications
Wang et al. A convolutional neural network method based on Adam optimizer with power-exponential learning rate for bearing fault diagnosis
CN106778846A (en) A kind of method for forecasting based on SVMs
Xia et al. Residual-hypergraph convolution network: A model-based and data-driven integrated approach for fault diagnosis in complex equipment
Guo et al. Improved adversarial learning for fault feature generation of wind turbine gearbox
CN113822418A (en) Wind power plant power prediction method, system, device and storage medium
CN112305388B (en) On-line monitoring and diagnosing method for insulation partial discharge faults of generator stator winding
CN116956215A (en) Fault diagnosis method and system for transmission system
Chatterjee et al. A dual transformer model for intelligent decision support for maintenance of wind turbines
Ruan et al. Remaining useful life prediction for aero-engine based on lstm and cnn
CN114298132A (en) Wind power prediction method and device and electronic equipment
Liu et al. Semi-supervised deep learning recognition method for the new classes of faults in wind turbine system
Liu et al. A novel wind turbine health condition monitoring method based on common features distribution adaptation

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
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20170531