CN109029975A - A kind of method for diagnosing faults of wind turbine gearbox - Google Patents
A kind of method for diagnosing faults of wind turbine gearbox Download PDFInfo
- Publication number
- CN109029975A CN109029975A CN201810667883.XA CN201810667883A CN109029975A CN 109029975 A CN109029975 A CN 109029975A CN 201810667883 A CN201810667883 A CN 201810667883A CN 109029975 A CN109029975 A CN 109029975A
- Authority
- CN
- China
- Prior art keywords
- neural network
- wind turbine
- turbine gearbox
- particle swarm
- diagnosing faults
- 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
- G01M13/021—Gearings
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Wind Motors (AREA)
Abstract
The method for diagnosing faults of wind turbine gearbox disclosed by the invention establishes fault set after decomposition firstly, extracting the vibration acceleration signal of wind turbine gearbox；Secondly, enhancing the robustness of quanta particle swarm optimization using a kind of random Adjusted Option towards converging diverging coefficient；Again, in order to further increase the probability that algorithm jumps out local optimum, a kind of restarting strategy is also introduced into quanta particle swarm optimization；Finally, establishing the fault diagnosis model of wind turbine gearbox using the method that improved quantum particle swarm is combined with BP neural network.Compared with the scheme of BP neural network, population and quantum telepotation BP network, method for diagnosing faults of the invention diagnostic accuracy with higher reduces severe accident odds.
Description
Technical field
The invention belongs to fault diagnosis of wind turbines technical fields, and in particular to a kind of fault diagnosis side of wind turbine gearbox
Method.
Background technique
Gear-box is one of critical component of Wind turbines.Due to being influenced by shock loading and alternating load, gear
The Frequent Troubles such as abrasion, the broken teeth of case, it has also become the multiple area of failure.According to statistics, gearbox fault accounts for about Wind turbines total failare
60%, be induce equipment fault an important factor for.Therefore, effective gearbox fault identification model is established, grasps it in time
Health status avoids severe accident odds, is conducive to mention predicted in advance the imminent failure of gear-box
The operation benefits of high wind power plant.
In wind turbine gearbox fault diagnosis technology, in analysis of vibration signal method using relatively broad.However, being based on
The conventional diagnostic technology of model is difficult to realize effective identification of gearbox fault mode, so artificial intelligence technology is introduced in wind
In the fault diagnosis of motor group, such as artificial neural network and support vector machines (SVM).Only artificial neural network often exists
The problems such as over-fitting, and the optimized parameter of SVM algorithm is difficult to obtain, and causes its diagnostic accuracy in urgent need to be improved.For this purpose, a series of
Intelligent optimization algorithm is applied wherein to obtain preferable parameter, as particle swarm algorithm (PSO), genetic algorithm (GA) and bacterium look for
Eat algorithm (BFA) etc..
In past 20 years, many intelligent optimization algorithms be used to solve Parametric optimization problem.Wherein, particle group optimizing
Algorithm is a kind of important swarm intelligence searching method.Since its concept is simple, is easily achieved, particle swarm algorithm receives evolution
The extensive concern of calculating field.However, PSO algorithm is easily trapped into local optimum, thus it cannot be guaranteed that global convergence.By quantum
Random behavior spy sexual enlightenment in mechanics, Sun Jun etc. propose quantum particle swarm optimization (QPSO), and demonstrate its overall situation
Convergence is better than PSO algorithm.In QPSO algorithm, each particle can be repaired by learning the optimum position information of entire population
The positive direction of search, has the characteristics that fast convergence rate, however, QPSO algorithm is also easy when solving complicated multiextremal optimization problem
Premature Convergence.
Summary of the invention
The object of the present invention is to provide a kind of method for diagnosing faults of wind turbine gearbox, solve existing artificial neural network
The problem of network over-fitting and convergence rate are slow, QPSO algorithm solution procedure easy Premature Convergence.
The technical scheme adopted by the invention is that a kind of method for diagnosing faults of wind turbine gearbox, specifically according to following step
It is rapid to implement:
Step 1, the vibration acceleration signal for acquiring gear surface, and after wavelet technique resolution process, obtain failure knowledge
This feature vector is randomly divided into two groups by another characteristic vector, respectively as the training set and test set of BP neural network；To tooth
The fault type of roller box is encoded, and as output sample；
Step 2, trains function and input layer to the biography of output layer at the hidden layer neuron number for determining BP neural network
Delivery function establishes the fault diagnosis model based on BP neural network；
Step 3 is sought using parameter of the improved quanta particle swarm optimization to the fault diagnosis model of BP neural network
It is excellent, optimal power threshold parameter is obtained, Optimized BP Neural Network is obtained；
Step 4 is trained Optimized BP Neural Network using training set, obtains improving quantum telepotation BP nerve
Network model；
Step 5 predicts test set using improvement quantum particle swarm Optimized BP Neural Network model, exports wind-powered electricity generation tooth
The diagnostic result of roller box failure, and the diagnostic result is compared with the output sample that step 1 obtains, judge wind power gear
The fault type of case.
The features of the present invention also characterized in that
In step 1, described eigenvector is respectively Power Spectral Entropy, Wavelet Entropy, correlation dimension, box counting dimension, kurtosis and the degree of bias.
In step 1, the fault type of the wind turbine gearbox includes that gear is normal, gear grinding undermines gear tooth breakage.
In step 2, the BP neural network uses three layers of BP neural network, and the number of hidden layer neuron is as follows
It is calculated:
In formula (1), s is hidden layer neuron number；M and n is respectively input, output layer neuron number, and m takes 6, n
Take 3；P is the integer between 1~10.
In step 2, the transmission function of the input layer and output layer is respectively tansig function and logsig function, training
Function selects trainlm function.
Step 3 specifically comprises the following steps:
Step 3.1, the improved quanta particle swarm optimization parameter of initialization；
The parameter of the improved quanta particle swarm optimization includes population scale N, maximum number of iterations kmax, adjust contraction-
3 parameters a, b and σ needed for flare factor β, limiting value limit, conversion parameter trial；
Step 3.2, random generation initial solution xi, calculate fitness fit (i), and enable current individual optimal solution pbi=xi, repeatedly
Generation number k=1；
Globally optimal solution gb and its fitness f in step 3.3, record populationbest；
If step 3.4, k < kmax, then follow the steps 3.5；Otherwise 3.9 are gone to step；
Step 3.5 generates converging diverging factor beta；
The create-rule of converging diverging factor beta are as follows:
μ=a+brand (0,1) (2),
β=μ+σ randn (0,1) (3)；
In formula (2) and (3), a, b and σ are 3 preset constants；Rand (0,1) is to obey equally distributed random number；
Randn (0,1) is the random number for obeying standardized normal distribution.
If step 3.6, trial < limit, new candidate solution is generated using the update rule of quanta particle swarm optimization；It is no
Then, new solution is generated using restarting strategy；
The quality for the solution that step 3.7, appraisal procedure 3.6 obtain；
If this solution is better than current individual optimal solution pb, substitution pb is solved with this, and enable conversion parameter trial=0；Otherwise
Trial=trial+1；If this solution is better than globally optimal solution gb, substitution gb is solved with this；
Step 3.8 enables the number of iterations k=k+1, and gos to step 3.4；
Step 3.9, using optimal solution gb as the power threshold parameter of BP neural network, obtain Optimized BP Neural Network.
In step 3.6, the update rule of the quanta particle swarm optimization are as follows:
In formula (4) and (5), u andIt is the random number between 0~1.
In step 3.6, the restarting strategy is indicated are as follows:
In formula (7), lb and ub are respectively lower bound and the upper bound of search space；Rand and rand (0,1) is between 0~1
Random number.
The beneficial effects of the present invention are:
A kind of method for diagnosing faults of wind turbine gearbox of the present invention, by the random regulation scheme of control parameter and restarts plan
It is slightly introduced into quanta particle swarm optimization, and a limiting value is set to control the implementation of restarting strategy, enhance quanta particle
Then the convergence precision and convergence rate of group's algorithm combine improved quanta particle swarm optimization with BP neural network, structure
A kind of improvement quantum particle swarm optimization neural network model is built, to find the optimal initial weight and threshold value ginseng of BP neural network
It counts, and identifies the fault type of wind turbine gearbox with this, diagnostic method of the invention reduces BP network and falls into local optimum
Risk, improve the accuracy of wind turbine gearbox failure modes, be conducive to improve wind power plant operation benefits.
Detailed description of the invention
Fig. 1 is a kind of flow chart of the method for diagnosing faults of wind turbine gearbox of the present invention；
Fig. 2 is the fitness convergence curve figure of tri- kinds of algorithms of PSOBP, QPSOBP and IQPSOBP；
Fig. 3 is the training error convergence curve figure of tri- kinds of algorithms of PSOBP, QPSOBP and IQPSOBP.
Specific embodiment
The following describes the present invention in detail with reference to the accompanying drawings and specific embodiments.
In order to improve the accuracy of wind power gear box fault diagnosis, the invention discloses a kind of failures of wind turbine gearbox to examine
Disconnected method establishes fault set after decomposition firstly, extracting the vibration acceleration signal of wind turbine gearbox；Secondly, using it is a kind of towards
The random Adjusted Option of converging diverging coefficient enhances the robustness of quanta particle swarm optimization；Again, it is calculated to further increase
Method jumps out the probability of local optimum, and a kind of restarting strategy is also introduced into quanta particle swarm optimization；Finally, using improved
The method that quantum particle swarm is combined with BP neural network establishes the fault diagnosis model of wind turbine gearbox, with BP nerve net
The scheme of network, population and quantum telepotation BP network is compared, and method for diagnosing faults of the invention is with higher to examine
Disconnected precision, reduces severe accident odds.
As shown in Figure 1, specific implement in accordance with the following steps:
Step 1, the vibration acceleration signal for acquiring gear surface, and after wavelet technique resolution process, obtain failure knowledge
Another characteristic vector, and this feature vector is randomly divided into two groups, respectively as the training set and test set of BP neural network；It is right
The fault type of gear-box is encoded, and as output sample.
Wherein, the feature vector of wind turbine gearbox fault identification is respectively Power Spectral Entropy, Wavelet Entropy, correlation dimension, box dimension
Number, kurtosis and the degree of bias.
The fault type of wind turbine gearbox includes gear normal, gear wear and gear tooth breakage, respectively with coding when experiment
[1 0 0], [0 1 0] and [0 0 1] indicate.
The transmitting of step 2, the hidden layer neuron number for determining BP neural network, training function, input layer and output layer
Function establishes the fault diagnosis model based on BP neural network.
It can be fitted arbitrarily complicated nonlinear system in view of three layers of BP neural network, so carried out using three layers of BP network
Fault diagnosis.According to the dimension of feature vector and fault type, input layer number is set as 6, output layer neuron number
It is set as 3.
In addition, determining the value range of hidden layer neuron number using empirical equation, determined further according to experimental result
Its occurrence.When experiment, selects different integers as hidden layer neuron number, record the output error of network, and will most
Integer corresponding to small error is as hidden layer neuron number.
Wherein, the number of hidden layer neuron is first calculated by following empirical equation:
In formula (1), s is hidden layer neuron number；M and n is respectively input, output layer neuron number；P is 1~10
Between integer.
The transmission function of input layer and hidden layer neuron is tansig, and the transmission function of output layer neuron is
Logsig, training function are trainlm.
Step 3 is sought using parameter of the improved quanta particle swarm optimization to the fault diagnosis model of BP neural network
It is excellent, optimal power threshold parameter is obtained, Optimized BP Neural Network is obtained, specifically comprises the following steps:
Step 3.1, the improved quanta particle swarm optimization parameter of initialization；
The parameter of the improved quanta particle swarm optimization includes population scale N, maximum number of iterations kmax, adjust contraction-
3 parameters a, b and σ needed for flare factor β, limiting value limit, conversion parameter trial；
Step 3.2, random generation initial solution xi, calculate fitness fit (i), and enable current individual optimal solution pbi=xi, repeatedly
Generation number k=1；
Globally optimal solution gb and its fitness f in step 3.3, record populationbest；
If step 3.4, k < kmax, then follow the steps 3.5；Otherwise 3.9 are gone to step；
Step 3.5 generates converging diverging factor beta；
The create-rule of converging diverging factor beta are as follows:
μ=a+brand (0,1) (2),
β=μ+σ randn (0,1) (3)；
In formula (2) and (3), a, b and σ are 3 preset constants；Rand (0,1) is to obey equally distributed random number；
Randn (0,1) is the random number for obeying standardized normal distribution.
If step 3.6, trial < limit, new candidate solution is generated using the update rule of quanta particle swarm optimization；It is no
Then, new solution is generated using restarting strategy；
Wherein, the update Rule Expression of quanta particle swarm optimization are as follows:
In formula (4) and (5), u andIt is the random number between 0~1.
Restarting strategy indicates are as follows:
In formula (7), lb and ub are respectively lower bound and the upper bound of search space；Rand and rand (0,1) is between 0~1
Random number.
Obviously, the top half in formula (7) is identical as the initial method of particle position, and the multiplicity of population can be enhanced
Property, however, frequent initialization inevitably reduces convergence speed of the algorithm, or even make it that can not converge to optimal value；
In addition, globally optimal solution has very strong guiding function in evolutionary process, its introducing can accelerate convergence rate.For reality
The letter of random initializtion and global optimum's individual is utilized in balance between existing convergence rate and population diversity, formula (7) simultaneously
Breath.
The quality for the solution that step 3.7, appraisal procedure 3.6 obtain；
The quality of this solution is determined according to fitness value.If this solution is better than current individual optimal solution pb, substituted with this solution
Pb, and enable conversion parameter trial=0；Otherwise trial=trial+1；If this solution is better than globally optimal solution gb, replaced with this solution
For gb.
Step 3.8 enables the number of iterations k=k+1, and gos to step 3.4；
Step 3.9, using optimal solution gb as the power threshold parameter of BP neural network, obtain Optimized BP Neural Network.
Step 4 is trained Optimized BP Neural Network using training set, obtains improving quantum telepotation BP nerve
Network model；
Step 5 predicts test set using improvement quantum particle swarm Optimized BP Neural Network model, exports wind-powered electricity generation tooth
The diagnostic result of roller box failure, and the diagnostic result is compared with the output sample that step 1 obtains, judge wind power gear
The fault type of case.
Diagnosis example
Using certain wind power plant 1.5MW Wind turbines as research object, gear-box is by primary planet wheel and two-stage parallel axis gear
Transmission composition.Pass through the acceleration transducer being mounted on high speed shaft of gearbox bearing block position parallel-axes gears table collected
The vibration acceleration signal in face obtains feature vector after decomposing, and constructs and improve quantum particle swarm BP neural network
(IQPSOBP) model carries out fault diagnosis to wind turbine gearbox.In addition, by its diagnostic result and BP neural network, population BP
The output result of neural network (PSOBP) and quantum particle swarm BP neural network (QPSOBP) compares.
According to the dimension of feature vector and fault type, 6 are set by the input layer number of BP neural network, output layer
In number of nodes be equal to 3, node in hidden layer is set as 10.In addition, for three layers of BP network, the transmission function of hidden layer neuron
For tansig, the transmission function of output layer neuron is logsig, and training function is trainlm, and frequency of training is set as 5000,
Learning rate is equal to 0.1, and the error convergence factor is set as 10E-5, limiting value limit=10.The kind of PSOBP, QPSOBP and IQPSOBP
Group's scale is disposed as 30, and maximum number of iterations is equal to 150.In addition, using based on BP network, PSOBP, QPSOBP and
5 groups of test datas are verified and are analyzed by the wind turbine gearbox fault identification model of IQPSOBP.
Table 1 gives the diagnostic result of four kinds of algorithms.
The result that table 1 diagnoses wind turbine gearbox failure using algorithms of different
Fig. 2 be tri- kinds of algorithms of PSOBP, QPSOBP and IQPSOBP fitness convergence curve figure, Fig. 3 PSOBP,
The training error convergence curve figure of tri- kinds of algorithms of QPSOBP and IQPSOBP, as can be seen that IQPSOBP algorithm from Fig. 2 and Fig. 3
With most fast convergence rate and highest convergence precision.
In addition, as can be seen from Table 1, according to default exports coding [1 0 0], [0 1 0] and [0 0 1], QPSOBP
It can correctly identify the fault mode of gear-box with IQPSOBP, but the diagnostic result of IQPSOBP model is closer to default volume
Code.Therefore, the wind turbine gearbox fault diagnosis model based on IQPSOBP algorithm has highest accuracy of identification, facilitates engineering
Practice.
Claims (8)
1. a kind of method for diagnosing faults of wind turbine gearbox, which is characterized in that specifically implement in accordance with the following steps:
Step 1, the vibration acceleration signal for acquiring gear surface, and after wavelet technique resolution process, obtain fault identification
This feature vector is randomly divided into two groups by feature vector, respectively as the training set and test set of BP neural network；To gear-box
Fault type encoded, and as output sample；
Step 2, trains function and input layer to the transmitting letter of output layer at the hidden layer neuron number for determining BP neural network
Number establishes the fault diagnosis model based on BP neural network；
Step 3 carries out optimizing using parameter of the improved quanta particle swarm optimization to the fault diagnosis model of BP neural network, obtains
Optimal power threshold parameter is obtained, Optimized BP Neural Network is obtained；
Step 4 is trained Optimized BP Neural Network using training set, obtains improving quantum particle swarm Optimized BP Neural Network
Model；
Step 5 predicts test set using improvement quantum particle swarm Optimized BP Neural Network model, exports wind turbine gearbox
The diagnostic result of failure, and the diagnostic result is compared with the output sample that step 1 obtains, judge wind turbine gearbox
Fault type.
2. a kind of method for diagnosing faults of wind turbine gearbox as described in claim 1, which is characterized in that in step 1, the spy
Levying vector is respectively Power Spectral Entropy, Wavelet Entropy, correlation dimension, box counting dimension, kurtosis and the degree of bias.
3. a kind of method for diagnosing faults of wind turbine gearbox as described in claim 1, which is characterized in that in step 1, the wind
The fault type of electrical gearbox includes that gear is normal, gear grinding undermines gear tooth breakage.
4. a kind of method for diagnosing faults of wind turbine gearbox as described in claim 1, which is characterized in that in step 2, the BP
Neural network uses three layers of BP neural network, and the number of hidden layer neuron is calculated as follows:
In formula (1), s is hidden layer neuron number；M and n is respectively input, output layer neuron number, and m takes 6, n to take 3；p
For the integer between 1~10.
5. a kind of method for diagnosing faults of wind turbine gearbox as described in claim 1, which is characterized in that described defeated in step 2
The transmission function for entering layer and output layer is respectively tansig function and logsig function, and training function selects trainlm function.
6. a kind of method for diagnosing faults of wind turbine gearbox as described in claim 1, which is characterized in that step 3 specifically includes
Following steps:
Step 3.1, the improved quanta particle swarm optimization parameter of initialization；
The parameter of the improved quanta particle swarm optimization includes population scale N, maximum number of iterations kmax, adjust converging diverging
3 parameters a, b and σ needed for factor beta, limiting value limit, conversion parameter trial；
Step 3.2, random generation initial solution xi, calculate fitness fit (i), and enable current individual optimal solution pbi=xi, iteration time
Number k=1；
Globally optimal solution gb and its fitness f in step 3.3, record populationbest；
If step 3.4, k < kmax, then follow the steps 3.5；Otherwise 3.9 are gone to step；
Step 3.5 generates converging diverging factor beta；
The create-rule of converging diverging factor beta are as follows:
μ=a+brand (0,1) (2),
β=μ+σ randn (0,1) (3)；
In formula (2) and (3), a, b and σ are 3 preset constants；Rand (0,1) is to obey equally distributed random number；randn
(0,1) random number to obey standardized normal distribution；
If step 3.6, trial < limit, new candidate solution is generated using the update rule of quanta particle swarm optimization；Otherwise, it adopts
New solution is generated with restarting strategy；
The quality for the solution that step 3.7, appraisal procedure 3.6 obtain；
If this solution is better than current individual optimal solution pb, substitution pb is solved with this, and enable conversion parameter trial=0；Otherwise trial
=trial+1；If this solution is better than globally optimal solution gb, substitution gb is solved with this；
Step 3.8 enables the number of iterations k=k+1, and gos to step 3.4；
Step 3.9, using optimal solution gb as the power threshold parameter of BP neural network, obtain Optimized BP Neural Network.
7. a kind of method for diagnosing faults of wind turbine gearbox as claimed in claim 6, which is characterized in that described in step 3.6
The update rule of quanta particle swarm optimization are as follows:
In formula (4) and (5), u andIt is the random number between 0~1.
8. a kind of method for diagnosing faults of wind turbine gearbox as claimed in claim 6, which is characterized in that described in step 3.6
Restarting strategy indicates are as follows:
In formula (7), lb and ub are respectively lower bound and the upper bound of search space；Rand and rand (0,1) be between 0~1 with
Machine number.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810667883.XA CN109029975B (en) | 2018-06-26 | 2018-06-26 | Fault diagnosis method for wind power gear box |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810667883.XA CN109029975B (en) | 2018-06-26 | 2018-06-26 | Fault diagnosis method for wind power gear box |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109029975A true CN109029975A (en) | 2018-12-18 |
| CN109029975B CN109029975B (en) | 2020-08-04 |
Family
ID=64610966
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810667883.XA Expired - Fee Related CN109029975B (en) | 2018-06-26 | 2018-06-26 | Fault diagnosis method for wind power gear box |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109029975B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109657647A (en) * | 2019-01-09 | 2019-04-19 | 红河学院 | A kind of Fault Diagnosis Method of Hydro-generating Unit based on information fusion technology |
| CN110044621A (en) * | 2019-03-25 | 2019-07-23 | 西安交通大学 | The epicyclic gearbox oscillation power of gear distress composes prediction technique |
| CN110119778A (en) * | 2019-05-10 | 2019-08-13 | 辽宁大学 | A kind of equipment method for detecting health status improving chicken group's optimization RBF neural |
| CN110988673A (en) * | 2019-11-05 | 2020-04-10 | 国网河北省电力有限公司电力科学研究院 | Motor rotor fault detection method and device and terminal equipment |
| CN111189638A (en) * | 2019-12-24 | 2020-05-22 | 沈阳化工大学 | HMM and QPSO optimization algorithm-based bearing fault degree identification method |
| CN111521952A (en) * | 2020-05-08 | 2020-08-11 | 红河学院 | Synchronous rectification circuit, charger and synchronous rectification circuit control method and system |
| CN112202196A (en) * | 2020-08-18 | 2021-01-08 | 广西大学 | Quantum deep reinforcement learning control method of doubly-fed wind generator |
| CN113063589A (en) * | 2021-03-18 | 2021-07-02 | 重庆青山工业有限责任公司 | Gear microscopic error vibration prediction method based on neural network |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101221213A (en) * | 2008-01-25 | 2008-07-16 | 湖南大学 | Analogue circuit fault diagnosis neural network method based on particle swarm algorithm |
| CN102969780A (en) * | 2012-10-30 | 2013-03-13 | 天津大学 | Off-grid wind/solar/battery hybrid power generation system capacity optimal configuration method |
| CN103278326A (en) * | 2013-06-14 | 2013-09-04 | 上海电机学院 | Method for diagnosing faults of wind generating set gear case |
| CN104050504A (en) * | 2013-03-11 | 2014-09-17 | 江南大学 | Chaotic-system parameter estimation method based on neighborhood quantum particle swarm |
| CN104458252A (en) * | 2014-08-30 | 2015-03-25 | 吉林大学 | Method for monitoring running state of high-speed train gear box |
| CN106199332A (en) * | 2016-06-29 | 2016-12-07 | 中国铁路总公司 | A kind of track circuit red band method for diagnosing faults of improvement |
| CN106548230A (en) * | 2016-10-14 | 2017-03-29 | 云南电网有限责任公司昆明供电局 | Diagnosis Method of Transformer Faults based on Modified particle swarm optimization neutral net |
| CN107546781A (en) * | 2017-09-06 | 2018-01-05 | 广西电网有限责任公司电力科学研究院 | Micro-capacitance sensor multiple target running optimizatin method based on PSO innovatory algorithms |
| CN107798379A (en) * | 2017-11-23 | 2018-03-13 | 东北大学 | Improve the method for quantum particle swarm optimization and the application based on innovatory algorithm |
| CN107995027A (en) * | 2017-11-23 | 2018-05-04 | 东北大学 | Improved quantum particle swarm optimization and the method applied to prediction network traffics |
-
2018
- 2018-06-26 CN CN201810667883.XA patent/CN109029975B/en not_active Expired - Fee Related
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101221213A (en) * | 2008-01-25 | 2008-07-16 | 湖南大学 | Analogue circuit fault diagnosis neural network method based on particle swarm algorithm |
| CN102969780A (en) * | 2012-10-30 | 2013-03-13 | 天津大学 | Off-grid wind/solar/battery hybrid power generation system capacity optimal configuration method |
| CN104050504A (en) * | 2013-03-11 | 2014-09-17 | 江南大学 | Chaotic-system parameter estimation method based on neighborhood quantum particle swarm |
| CN103278326A (en) * | 2013-06-14 | 2013-09-04 | 上海电机学院 | Method for diagnosing faults of wind generating set gear case |
| CN104458252A (en) * | 2014-08-30 | 2015-03-25 | 吉林大学 | Method for monitoring running state of high-speed train gear box |
| CN106199332A (en) * | 2016-06-29 | 2016-12-07 | 中国铁路总公司 | A kind of track circuit red band method for diagnosing faults of improvement |
| CN106548230A (en) * | 2016-10-14 | 2017-03-29 | 云南电网有限责任公司昆明供电局 | Diagnosis Method of Transformer Faults based on Modified particle swarm optimization neutral net |
| CN107546781A (en) * | 2017-09-06 | 2018-01-05 | 广西电网有限责任公司电力科学研究院 | Micro-capacitance sensor multiple target running optimizatin method based on PSO innovatory algorithms |
| CN107798379A (en) * | 2017-11-23 | 2018-03-13 | 东北大学 | Improve the method for quantum particle swarm optimization and the application based on innovatory algorithm |
| CN107995027A (en) * | 2017-11-23 | 2018-05-04 | 东北大学 | Improved quantum particle swarm optimization and the method applied to prediction network traffics |
Non-Patent Citations (5)
| Title |
|---|
| 李宝磊 等: "多元优化算法及其收敛性分析", 《自动化学报》 * |
| 李敏 等: "基于改进量子粒子群优化模糊聚类的变压器故障诊断方法", 《电测与仪表》 * |
| 程加堂 等: "CQPSO-BP 算法在风电机组齿轮箱故障诊断中的应用", 《太阳能学报》 * |
| 马欢 等: "空间机器人惯性参数辨识的粒子群优化新算法", 《宇航学报》 * |
| 龙泉 等: "基于粒子群优化BP神经网络的风电机组齿轮箱故障诊断方法", 《太阳能学报》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109657647A (en) * | 2019-01-09 | 2019-04-19 | 红河学院 | A kind of Fault Diagnosis Method of Hydro-generating Unit based on information fusion technology |
| CN110044621A (en) * | 2019-03-25 | 2019-07-23 | 西安交通大学 | The epicyclic gearbox oscillation power of gear distress composes prediction technique |
| CN110119778A (en) * | 2019-05-10 | 2019-08-13 | 辽宁大学 | A kind of equipment method for detecting health status improving chicken group's optimization RBF neural |
| CN110119778B (en) * | 2019-05-10 | 2024-01-05 | 辽宁大学 | Equipment health state detection method for improving chicken flock optimization RBF neural network |
| CN110988673A (en) * | 2019-11-05 | 2020-04-10 | 国网河北省电力有限公司电力科学研究院 | Motor rotor fault detection method and device and terminal equipment |
| CN111189638A (en) * | 2019-12-24 | 2020-05-22 | 沈阳化工大学 | HMM and QPSO optimization algorithm-based bearing fault degree identification method |
| CN111189638B (en) * | 2019-12-24 | 2021-08-06 | 沈阳化工大学 | HMM and QPSO optimization algorithm-based bearing fault degree identification method |
| CN111521952A (en) * | 2020-05-08 | 2020-08-11 | 红河学院 | Synchronous rectification circuit, charger and synchronous rectification circuit control method and system |
| CN112202196A (en) * | 2020-08-18 | 2021-01-08 | 广西大学 | Quantum deep reinforcement learning control method of doubly-fed wind generator |
| CN113063589A (en) * | 2021-03-18 | 2021-07-02 | 重庆青山工业有限责任公司 | Gear microscopic error vibration prediction method based on neural network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109029975B (en) | 2020-08-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109029975A (en) | A kind of method for diagnosing faults of wind turbine gearbox | |
| Zhang et al. | The multiclass fault diagnosis of wind turbine bearing based on multisource signal fusion and deep learning generative model | |
| CN106124212B (en) | Fault Diagnosis of Roller Bearings based on sparse coding device and support vector machines | |
| CN107449603A (en) | Fault Diagnosis of Fan method based on SVMs | |
| CN103033359B (en) | A kind of main transmission in wind generating set method for diagnosing faults of multiple features Multilateral Comprehensive Judge | |
| CN104751229B (en) | Bearing fault diagnosis method capable of recovering missing data of back propagation neural network estimation values | |
| CN109213121B (en) | Method for diagnosing clamping cylinder fault of fan braking system | |
| CN108304941A (en) | A kind of failure prediction method based on machine learning | |
| CN1811367A (en) | Integrated supporting vector machine mixed intelligent diagnosing method for mechanical fault | |
| CN116010900A (en) | Multi-scale feature fusion gearbox fault diagnosis method based on self-attention mechanism | |
| CN112784920B (en) | Yun Bianduan coordinated rotating component reactance domain self-adaptive fault diagnosis method | |
| CN107563251A (en) | Fault Diagnosis of Fan method based on extreme learning machine | |
| CN112945552B (en) | Gear fault diagnosis method based on variable-node double-hidden-layer extreme learning machine | |
| CN117290800B (en) | Timing sequence anomaly detection method and system based on hypergraph attention network | |
| Fan et al. | Deep mixed domain generalization network for intelligent fault diagnosis under unseen conditions | |
| Huang et al. | Deep residual networks-based intelligent fault diagnosis method of planetary gearboxes in cloud environments | |
| CN111678679A (en) | Circuit breaker fault diagnosis method based on PCA-BPNN | |
| Tao et al. | Multi-signal fusion diagnosis of gearbox based on minimum Bayesian risk reclassification and adaptive weighting | |
| Tyagi et al. | A hybrid genetic algorithm and back-propagation classifier for gearbox fault diagnosis | |
| Liang et al. | 1d convolutional neural networks for fault diagnosis of high-speed train bogie | |
| Kavianpour et al. | An intelligent gearbox fault diagnosis under different operating conditions using adversarial domain adaptation | |
| Vo et al. | Harnessing attention mechanisms in a comprehensive deep learning approach for induction motor fault diagnosis using raw electrical signals | |
| CN116383757B (en) | Bearing fault diagnosis method based on multi-scale feature fusion and migration learning | |
| CN116797204A (en) | Primary air quantity fault early warning method for coal mill based on wavelet decomposition and reconstruction and TCN-GRU-Self-Attention | |
| CN109214103A (en) | The reliability of gears analysis system of confrontation network is generated based on boundary constraint |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200804 Termination date: 20210626 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |