CN109615109A - Deep learning wind-powered electricity generation warning information analysis method based on Small Sample Database - Google Patents
Deep learning wind-powered electricity generation warning information analysis method based on Small Sample Database Download PDFInfo
- Publication number
- CN109615109A CN109615109A CN201811332363.XA CN201811332363A CN109615109A CN 109615109 A CN109615109 A CN 109615109A CN 201811332363 A CN201811332363 A CN 201811332363A CN 109615109 A CN109615109 A CN 109615109A
- Authority
- CN
- China
- Prior art keywords
- neuron
- sample
- indicate
- data
- probability
- 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.)
- Withdrawn
Links
- 230000005611 electricity Effects 0.000 title claims abstract description 16
- 238000004458 analytical method Methods 0.000 title claims abstract description 15
- 238000013135 deep learning Methods 0.000 title claims abstract description 11
- 238000012549 training Methods 0.000 claims abstract description 12
- 238000013528 artificial neural network Methods 0.000 claims abstract description 10
- 238000000605 extraction Methods 0.000 claims abstract description 7
- 210000002569 neuron Anatomy 0.000 claims description 54
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 claims description 3
- 230000001186 cumulative effect Effects 0.000 claims description 3
- 238000009499 grossing Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 210000005036 nerve Anatomy 0.000 claims description 3
- 230000001537 neural effect Effects 0.000 claims description 3
- 230000007257 malfunction Effects 0.000 abstract description 4
- 238000007405 data analysis Methods 0.000 abstract description 2
- 238000000034 method Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000007812 deficiency Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Educational Administration (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The deep learning wind-powered electricity generation warning information analysis method based on Small Sample Database that the invention discloses a kind of, by the characteristic parameter, wind speed and the output power that extract blower each section component, as training sample, using the limited Boltzmann machine model of contrast divergence algorithm training, Characteristic Extraction is carried out to data, then the characteristic quantity of extraction is subjected to statistical classification with probabilistic neural network algorithm, mark off corresponding failure risk probability, then early warning is carried out according to failure risk grade, realizes the fault warning of Wind turbines.The present invention carries out Characteristic Extraction to data sample by limited Boltzmann machine, plays the role of dimensionality reduction, provides good basis for the analysis of subsequent data, reduces the complexity of data analysis；And the problem of overcoming local minimum improves the accuracy of probability of malfunction risk class.
Description
Technical field
It is based on depth Boltzmann machine and probability mind the present invention relates to a kind of wind power plant fault warning information analysis method
Probability of malfunction risk is divided through network algorithm, to improve the precision of failure risk early warning, and in particular to a kind of
Deep learning wind-powered electricity generation warning information analysis method based on Small Sample Database.
Background technique
With the fast development of wind energy and putting into operation for large-scale wind power unit, and since most of units' installation is inclined
All there is operation troubles, directly affect wind-power electricity generation in remote area, factors, many Wind turbines in China such as load is unstable
Safety and economy.For the long-term stability development for keeping wind-powered electricity generation, enhance the competitiveness of it and traditional energy, it is necessary to constantly reduce
The cost of wind-power electricity generation.
Since China's wind-power electricity generation cause is started late, the failure risk study of warning of Wind turbines is also in primary rank
Section realizes that effective early warning of Wind turbines failure is the main problem faced now, Wind turbines failure risk mainly with electricity
Gas system is related to control system, and related to the external environments such as wind speed, and above data type is different and correlation is complicated.Mesh
Preceding common fault early warning method has artificial neural network, and fuzzy set theory, evidence theory is although method listed above can
The fault message of diagnosis uncertainty is calculated with different algorithms, but its conclusion all differs greatly with actual conditions.
Summary of the invention
The technical problems to be solved by the present invention are: overcoming the deficiencies of the prior art and provide, a kind of fault-tolerance is strong, reduces
Data complicated degree of analysis and the deep learning wind-powered electricity generation alarm based on Small Sample Database for improving probability of malfunction risk class accuracy
Information analysis method.
The technical scheme is that
A kind of deep learning wind-powered electricity generation warning information analysis method based on Small Sample Database, comprising the following steps:
Step 1, the data information that wind power plant detects is pre-processed first, obtains training sample.
Step 2, using limited Boltzmann machine, Characteristic Extraction is carried out to sample data, selects n key feature
Amount.
Step 3, using the data sample comprising n key feature amount extracted as training set, it is input to probabilistic neural
It in network, is trained, establishes the model of a display failure risk probability.
Step 4, the failure risk probability mould new data sample input step 3 comprising n key feature amount obtained
In type, failure risk probability is obtained, early warning is then carried out according to risk class.
In the above-mentioned deep learning wind-powered electricity generation warning information analysis method based on Small Sample Database, the realization packet of step 2
It includes:
Step 2.1, data wind power plant detected after normalized, obtain the matrix of a p ' n rank:, wherein have n sample, p characteristic quantity；
(1);
Step 2.2, the limited Boltzmann machine comprising p visible layer neuron and m hidden layer neuron is constructed, it is assumed that
Visible layer neuron and hidden layer neuron obey Bernoulli Jacob's distribution, each visible layer neuron node and hidden layer nerve
It is as follows that energy value between first node meets formula:
(2)；
vIndicate all visible layer neurons,hIndicate all hidden layer neurons,Indicate i-th of visible neuronal and jth
Big weight between a hidden neuron,Indicate the offset threshold of i-th of visible layer neuron,It indicates to hide mind j-th
Offset threshold through member,Indicate the parameter (being real number) of limited Boltzmann machine model.
Step 2.3, in given visible layer, irrelevant between each hidden layer neuron value, probability distribution is such as
Following formula:
(3)；
In given hidden layer, also irrelevant between all visible layer neuron values, probability distribution is as follows:
(4)；
Step 2.4, for the X in training set, X is attached to visible layer, calculates the probability that hidden layer neuron is turned on, as follows:
(5)；
In formula, the meaning of 0 and 1 state here is to represent model to choose which node comes using when being active
Value is 1, and the state value that is not activated is 0.
Step 2.5, a sample is extracted from calculated probability distribution, i.e.,
(6)；
Visible layer is reconstructed with h, i.e.,
(7)；
Equally, a sample of visible layer is extracted, i.e.,
(8)；
The probability that hidden layer neuron is turned on is calculated with visible layer neuron (after reconstruct) again, i.e.,
(9)；
Step 2.6, weight is updated by formula (10), until error rate reaches minimum, to extract n key feature vector
(10)；
The realization of step 3 includes:
Step 3.1, the sample comprising n feature vector extracted is input in probabilistic neural network, is asked by formula (11)
Obtain the matching degree in each neuron and input layer in mode layer between each neuron；
(11)；
Wherein,l gIndicate the number of g class neuron；The number of n expression feature;Indicate smoothing parameter;Indicate the of g class
J-th of data of i neuron.
Step 3.2, it then by the cumulative summation of the matching degree of every class, then is averaged, obtains the failure wind of input sample
Dangerous probability.
The present invention has as follows a little: 1, carrying out Characteristic Extraction to data sample by limited Boltzmann machine, play
The effect of dimensionality reduction provides good basis for the analysis of subsequent data, reduces the complexity of data analysis.2, using mentioning
The characteristic quantity of taking-up is trained probabilistic neural network, and training is easy, fast convergence rate, is suitble to the place for real time data
Reason, the number of each layer neuron is relatively more fixed, is easy to hardware implementation.3, the Nonlinear Mapping letter of the radial base of hidden layer use simultaneously
Number, it is contemplated that staggeredly influencing between different classes of sample has very strong fault-tolerance, and overcome asking for local minimum
Topic, improves the accuracy of probability of malfunction risk class.Expansion performance is good, and network learning procedure is simple, fast convergence rate.
Detailed description of the invention
Fig. 1 is work flow diagram of the invention.
Fig. 2 is the schematic diagram that Characteristic Extraction is carried out with limited Boltzmann machine.
Fig. 3 is the schematic diagram that failure risk probability is obtained with probabilistic neural network.
Specific embodiment:
Below with reference to the embodiments and with reference to the accompanying drawing the technical solutions of the present invention will be further described.
Examples of the embodiments are shown in the accompanying drawings, and in which the same or similar labels are throughly indicated identical or classes
As element or element with the same or similar functions.The embodiments described below with reference to the accompanying drawings are exemplary, only
It is used to explain the present invention, and is not construed as limiting the claims.
Following disclosure provides many different embodiments or example is used to realize different structure of the invention.For letter
Change disclosure of the invention, hereinafter the component of specific examples and setting are described.They are merely examples, and purpose is not
It is to limit the present invention.In addition, the present invention can in different examples repeat reference numerals and/or letter.This repetition be for
Simplified and clear purpose, itself do not indicate discussed various embodiments and/or be arranged between relationship.In addition, this hair
It is bright provide the example of various specific techniques and material, but those of ordinary skill in the art may be aware that other techniques
The use of applicability and/or other materials.In addition, structure of the fisrt feature described below in the "upper" of second feature can be with
Be formed as the embodiment directly contacted including the first and second features, also may include that other feature is formed in first and second
Embodiment between feature, such first and second feature may not be direct contact.
The present embodiment realizes that the deep learning wind-powered electricity generation warning information based on Small Sample Database divides using following technical scheme
Analysis method, comprising the following steps:
Step 1, the data information that wind power plant detects is pre-processed first, obtains training sample.
Step 2, using depth Boltzmann machine, characteristic quantity is carried out to the pretreated data sample that step 1 obtains and is mentioned
It takes, to extract n key feature amount.
Step 3, using the data of n key feature amount obtained in step 2 as training set, it is input to probabilistic neural network
In, it is trained, establishes the model of a display failure risk probability.
Step 4, the probabilistic neural network risk mould new sample input step 3 comprising n key feature amount obtained
In type, then obtained failure risk probability makes fault pre-alarming according to risk class.
Further, the realization of step 2 includes:
Step 2.1, data wind power plant detected after normalized, obtain the matrix of a p ' n rank:, wherein have n sample, p characteristic quantity；
(1);
Step 2.2, the limited Boltzmann machine comprising p visible layer units and m hiding layer units is constructed, it is seen that layer list
First and hiding layer unit obeys Bernoulli Jacob's distribution, between each visible layer neuron node and hidden layer neuron node
It is as follows that energy value meets formula:
(2)；
V indicates that all visible layer neurons, h indicate all hidden layer neurons,Indicate i-th of visible layer neuron and
Weight between j hidden layer neuron,Indicate the offset threshold of i-th of visible layer neuron,Indicate j-th of hidden layer
The offset threshold of neuron,Indicate the parameter of limited Boltzmann machine model.
Step 2.3, in given visible layer, irrelevant, probability distribution between each hidden layer neuron value
It is as follows:
(3)；
In given hidden layer, also irrelevant between all visible layer neuron values, probability distribution is as follows:
(4)；
Step 2.4, for the X in training set, X is attached to visible layer, the probability that hidden layer neuron is turned on is calculated, such as formula
(5):
(5)；
In formula, k indicates the number of sample, and the meaning of 0 and 1 state here is to represent model to choose which node to make
With being active duration is 1, and the state value that is not activated is 0.
Step 2.5, a sample is extracted from calculated probability distribution, i.e.,
(6)；
WithVisible layer is reconstructed, i.e.,
(7)；
Equally, a sample of visible layer is extracted, i.e.,
(8)；
The probability that hidden layer neuron is turned on is calculated with visible layer neuron (after reconstruct) again, i.e.,
(9)；
Step 2.6, weight is updated by formula (10), until error rate reaches minimum, to extract n key feature vector
(10)；
The realization of step 3 includes:
Step 3.1, the sample comprising n feature vector extracted is input in probabilistic neural network, is asked by formula (11)
Obtain the matching degree in each neuron and input layer in mode layer between each neuron.
(11)；
Wherein, lg indicates the number of g class neuron；The number of n expression feature;Indicate smoothing parameter;Indicate the of g class
J-th of data of i neuron.
Step 3.2, it then by the cumulative summation of the matching degree of every class, then is averaged, obtains the failure wind of input sample
Dangerous probability.
It should be understood that the part that this specification does not elaborate belongs to the prior art.
Although being described in conjunction with the accompanying a specific embodiment of the invention above, those of ordinary skill in the art should
Understand, these are merely examples, various deformation or modification can be made to these embodiments, without departing from original of the invention
Reason and essence.The scope of the present invention is only limited by the claims that follow.
Claims (3)
1. a kind of deep learning wind-powered electricity generation warning information analysis method based on Small Sample Database, characterized in that the following steps are included:
Step 1, the data information that wind power plant detects is pre-processed first, obtains training sample；
Step 2, using limited Boltzmann machine, Characteristic Extraction is carried out to sample data, selects n key feature amount；
Step 3, using the data sample comprising n key feature amount extracted as training set, it is input to probabilistic neural network
In, it is trained, establishes the model of a display failure risk probability；
Step 4, in the failure risk probabilistic model new data sample input step 3 comprising n key feature amount obtained,
Failure risk probability is obtained, early warning is then carried out according to risk class.
2. the deep learning wind-powered electricity generation warning information analysis method based on Small Sample Database as described in claim 1, it is characterized in that:
The realization of step 2 includes:
Step 2.1, data wind power plant detected after normalized, obtain the matrix of a p ' n rank:, wherein have n sample, p characteristic quantity；
(1);
Step 2.2, the limited Boltzmann machine comprising p visible layer neuron and m hidden layer neuron is constructed, it is assumed that
Visible layer neuron and hidden layer neuron obey Bernoulli Jacob's distribution, each visible layer neuron node and hidden layer nerve
It is as follows that energy value between first node meets formula:
(2)；
vIndicate all visible layer neurons,hIndicate all hidden layer neurons,Indicate i-th of visible neuronal and j-th
Big weight between hidden neuron,Indicate the offset threshold of i-th of visible layer neuron,It indicates to hide nerve j-th
The offset threshold of member,Indicate the parameter (being real number) of limited Boltzmann machine model；
Step 2.3, in given visible layer, irrelevant, probability distribution such as following formula between each hidden layer neuron value:
(3)；
In given hidden layer, also irrelevant between all visible layer neuron values, probability distribution is as follows:
(4)；
Step 2.4, for the X in training set, X is attached to visible layer, calculates the probability that hidden layer neuron is turned on, as follows:
(5)；
In formula, the meaning of 0 and 1 state here is to represent model to choose which node comes using when being active
Value is 1, and the state value that is not activated is 0；
Step 2.5, a sample is extracted from calculated probability distribution, i.e.,
(6)；
Visible layer is reconstructed with h, i.e.,
(7)；
Equally, a sample of visible layer is extracted, i.e.,
(8)；
The probability that hidden layer neuron is turned on is calculated with visible layer neuron (after reconstruct) again, i.e.,
(9)；
Step 2.6, weight is updated by formula (10), until error rate reaches minimum, to extract n key feature vector
(10).
3. as described in claim 1 in the deep learning wind-powered electricity generation warning information analysis method based on Small Sample Database, feature
Be: the realization of step 3 includes:
Step 3.1, the sample comprising n feature vector extracted is input in probabilistic neural network, is asked by formula (11)
Obtain the matching degree in each neuron and input layer in mode layer between each neuron；
(11)；
Wherein,l gIndicate the number of g class neuron；The number of n expression feature;Indicate smoothing parameter;Indicate the i-th of g class
J-th of data of a neuron；
Step 3.2, it then by the cumulative summation of the matching degree of every class, then is averaged, the failure risk for obtaining input sample is general
Rate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811332363.XA CN109615109A (en) | 2018-12-29 | 2018-12-29 | Deep learning wind-powered electricity generation warning information analysis method based on Small Sample Database |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811332363.XA CN109615109A (en) | 2018-12-29 | 2018-12-29 | Deep learning wind-powered electricity generation warning information analysis method based on Small Sample Database |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109615109A true CN109615109A (en) | 2019-04-12 |
Family
ID=66004034
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811332363.XA Withdrawn CN109615109A (en) | 2018-12-29 | 2018-12-29 | Deep learning wind-powered electricity generation warning information analysis method based on Small Sample Database |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109615109A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112766702A (en) * | 2021-01-13 | 2021-05-07 | 广东能源集团科学技术研究院有限公司 | Distributed power station fault analysis method and system based on deep belief network |
| CN113822421A (en) * | 2021-10-14 | 2021-12-21 | 平安科技（深圳）有限公司 | Neural network based anomaly positioning method, system, equipment and storage medium |
| CN113902745A (en) * | 2021-12-10 | 2022-01-07 | 山东捷瑞数字科技股份有限公司 | Method and device for identifying accurate fault of gearbox of commercial vehicle based on image processing |
| CN116471198A (en) * | 2023-06-19 | 2023-07-21 | 南京典格通信科技有限公司 | Power amplifier fault prediction method based on limited Boltzmann machine |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011014037A (en) * | 2009-07-03 | 2011-01-20 | Fuji Heavy Ind Ltd | Risk prediction system |
| US20140330762A1 (en) * | 2004-01-06 | 2014-11-06 | Neuric Llc | Electronic brain model with neuron tables |
| CN107730040A (en) * | 2017-09-30 | 2018-02-23 | 国网山东省电力公司电力科学研究院 | Power information system log information comprehensive characteristics extracting method and device based on RBM |
| CN108170994A (en) * | 2018-01-29 | 2018-06-15 | 河海大学 | A kind of oil-immersed electric reactor method for diagnosing faults based on two-way depth network |
| CN108898247A (en) * | 2018-06-22 | 2018-11-27 | 国网湖南省电力有限公司 | A kind of power grid Rainfall Disaster Risk Forecast Method, system and storage medium |
-
2018
- 2018-12-29 CN CN201811332363.XA patent/CN109615109A/en not_active Withdrawn
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140330762A1 (en) * | 2004-01-06 | 2014-11-06 | Neuric Llc | Electronic brain model with neuron tables |
| JP2011014037A (en) * | 2009-07-03 | 2011-01-20 | Fuji Heavy Ind Ltd | Risk prediction system |
| CN107730040A (en) * | 2017-09-30 | 2018-02-23 | 国网山东省电力公司电力科学研究院 | Power information system log information comprehensive characteristics extracting method and device based on RBM |
| CN108170994A (en) * | 2018-01-29 | 2018-06-15 | 河海大学 | A kind of oil-immersed electric reactor method for diagnosing faults based on two-way depth network |
| CN108898247A (en) * | 2018-06-22 | 2018-11-27 | 国网湖南省电力有限公司 | A kind of power grid Rainfall Disaster Risk Forecast Method, system and storage medium |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112766702A (en) * | 2021-01-13 | 2021-05-07 | 广东能源集团科学技术研究院有限公司 | Distributed power station fault analysis method and system based on deep belief network |
| CN113822421A (en) * | 2021-10-14 | 2021-12-21 | 平安科技（深圳）有限公司 | Neural network based anomaly positioning method, system, equipment and storage medium |
| CN113822421B (en) * | 2021-10-14 | 2024-05-14 | 平安科技（深圳）有限公司 | Neural network-based anomaly locating method, system, equipment and storage medium |
| CN113902745A (en) * | 2021-12-10 | 2022-01-07 | 山东捷瑞数字科技股份有限公司 | Method and device for identifying accurate fault of gearbox of commercial vehicle based on image processing |
| CN116471198A (en) * | 2023-06-19 | 2023-07-21 | 南京典格通信科技有限公司 | Power amplifier fault prediction method based on limited Boltzmann machine |
| CN116471198B (en) * | 2023-06-19 | 2023-10-03 | 南京典格通信科技有限公司 | Power amplifier fault prediction method based on limited Boltzmann machine |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109615109A (en) | Deep learning wind-powered electricity generation warning information analysis method based on Small Sample Database | |
| Tong et al. | An efficient deep model for day-ahead electricity load forecasting with stacked denoising auto-encoders | |
| Wang et al. | Echo state network based ensemble approach for wind power forecasting | |
| Kassa et al. | Short term wind power prediction using ANFIS | |
| CN108897954A (en) | Wind turbines temperature pre-warning method and its system based on BootStrap confidence calculations | |
| CN106408687B (en) | A kind of automobile engine fault early warning method based on machine learning method | |
| Ramkumar et al. | A Short‐Term Solar Photovoltaic Power Optimized Prediction Interval Model Based on FOS‐ELM Algorithm | |
| Cui et al. | An algorithm for forecasting day-ahead wind power via novel long short-term memory and wind power ramp events | |
| CN109444740A (en) | A kind of the malfunction intellectual monitoring and diagnostic method of Wind turbines | |
| CN106499656B (en) | A kind of fan wind speed intelligent control method | |
| CN109615147A (en) | A kind of following 72 hours air pollution forecasting method for early warning | |
| CN106483405A (en) | The method for diagnosing faults of the NPC photovoltaic DC-to-AC converter based on hidden Markov model | |
| Ge et al. | Short-term load forecasting of regional distribution network based on generalized regression neural network optimized by grey wolf optimization algorithm | |
| CN110009135A (en) | A kind of wind power forecasting method based on width study | |
| CN106371316A (en) | PSO-LSSVM-based on-line control method and apparatus for dosing of water island | |
| Huang et al. | Priori-guided and data-driven hybrid model for wind power forecasting | |
| Ningsih et al. | Wind speed forecasting using recurrent neural networks and long short term memory | |
| Liu et al. | Wind turbine fault detection with multimodule feature extraction network and adaptive strategy | |
| Yousefi et al. | An artificial neural network hybrid with wavelet transform for short-term wind speed forecasting: A preliminary case study | |
| Padmaja et al. | Stability and Reliability Analysis for Multiple WT Using Deep Reinforcement Learning | |
| Wang et al. | SVM based imbalanced correction method for Power Systems Transient stability evaluation | |
| Zhenhao et al. | Prediction of wind power ramp events based on deep neural network | |
| Liu et al. | Advanced evaluation method for regional wind power prediction | |
| Han et al. | Interpretation of Stability Assessment Machine Learning Models Based on Shapley Value | |
| Wang et al. | Prediction of wheat stripe rust based on neural networks |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20190412 |