CN110276468A - Method and device for predicting generated power of wind generating set - Google Patents
Method and device for predicting generated power of wind generating set Download PDFInfo
- Publication number
- CN110276468A CN110276468A CN201810210270.3A CN201810210270A CN110276468A CN 110276468 A CN110276468 A CN 110276468A CN 201810210270 A CN201810210270 A CN 201810210270A CN 110276468 A CN110276468 A CN 110276468A
- Authority
- CN
- China
- Prior art keywords
- image data
- prediction
- day image
- class
- generating set
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000003062 neural network model Methods 0.000 claims description 37
- PEDCQBHIVMGVHV-UHFFFAOYSA-N Glycerine Chemical compound OCC(O)CO PEDCQBHIVMGVHV-UHFFFAOYSA-N 0.000 claims description 36
- 238000005259 measurement Methods 0.000 claims description 20
- 238000010248 power generation Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 8
- 230000005611 electricity Effects 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 5
- 230000006870 function Effects 0.000 claims description 3
- 230000007547 defect Effects 0.000 abstract 1
- 230000001419 dependent effect Effects 0.000 abstract 1
- 238000004422 calculation algorithm Methods 0.000 description 34
- 238000010606 normalization Methods 0.000 description 10
- 230000008569 process Effects 0.000 description 7
- 241001269238 Data Species 0.000 description 6
- 230000007246 mechanism Effects 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000003064 k means clustering Methods 0.000 description 3
- 230000003595 spectral effect Effects 0.000 description 3
- 230000007812 deficiency Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000002068 genetic effect Effects 0.000 description 2
- 238000012886 linear function Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 210000005036 nerve Anatomy 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
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Supply And Distribution Of Alternating Current (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a method and equipment for predicting the generated power of a wind generating set. The method comprises the steps of establishing a generating power prediction model according to historical astronomical phenomena data of the position of the wind generating set and historical generating power of the wind generating set, wherein the generating power prediction model indicates the corresponding relation between the astronomical phenomena data and the generating power; and predicting the generated power of the wind generating set by using the generated power prediction model according to the predicted astronomical phenomena data. According to the method for predicting the generated power of the wind generating set, disclosed by the embodiment of the invention, aiming at the defects of the existing prediction method, the accuracy of data (weather data) which is dependent on the existing prediction and is used as a main factor for causing inaccurate generation power prediction is not high, and the accuracy of the generation power prediction is improved by replacing the input data of the prediction model with the astronomical phenomena data with higher accuracy.
Description
Technical field
The present invention relates to wind power generation fields, more particularly, are related to a kind of the pre- of the generated output of wind power generating set
Survey method and apparatus.
Background technique
With the development of wind generating technology, the continuous expansion of wind-powered electricity generation single-machine capacity and grid-connected wind farms scale, wind-powered electricity generation
The ratio for accounting for electric system power generation total amount also increases year by year.Wind power plant penetrates power and continues to increase, and gives electric system bring one
Series of problems becomes increasingly conspicuous, safety, stability and the reliability of serious prestige rib electric system.Wind power is carried out quasi- in time
True prediction can significantly increase safety, stability and the controllability of electric system.
The prediction technique of the generated output of current wind power generating set generally requires first acquisition numerical weather forecast system
It predicts the meteorological datas such as obtained wind speed, wind direction, the temperature of wind power plant, obtains wind power generating set using different technological means
Hub height at the data such as wind speed, wind direction, then predict wind power generating set using the power curve of wind power generating set
Output power；Or a kind of mapping relations are established between history meteorological data and Power Output for Wind Power Field and are predicted.By
In the meteorological datas such as wind speed, wind direction, temperature be by sensor measurement or according to weather forecast obtain, and measurement method itself with
And the accuracy of weather forecast has larger impact to the reliability of meteorological data, leads to the power generation function of existing wind power generating set
The accuracy of the prediction technique of rate is not high.
Summary of the invention
The purpose of the present invention is to provide a kind of prediction technique of the generated output of wind power generating set and equipment, to solve
The not high problem of the accuracy of existing prediction technique.
An aspect of of the present present invention provides a kind of prediction technique of the generated output of wind power generating set, comprising: according to wind-force
The history day image data of generating set position and the history generated output of wind power generating set establish generated power forecasting mould
Type, wherein generated power forecasting model indicates the corresponding relationship between day image data and generated output；According to the astronomical phenomena number of prediction
According to by using the generated output of generated power forecasting model prediction wind power generating set.
Optionally, prediction technique further include: before establishing generated power forecasting model, history day image data is gathered
History day image data is divided into multiple classes by class, wherein generated power forecasting model includes multiple neural network models, often
A neural network model corresponds to a class in the multiple class, wherein the step of establishing generated power forecasting model include:
Neural network model corresponding with each class is established according to the history day image data of each class and history generated output.
Optionally, the step of predicting the generated output of wind power generating set comprises determining that belonging to the day image data of prediction
Class, according to the day image data of prediction, by using the corresponding neural network model of class belonging to the day image data with prediction, prediction
The generated output of wind power generating set.
Optionally it is determined that the step of class belonging to the day image data of prediction, comprises determining that the cluster centre of each class and pre-
The similarity of the day image data of survey, and the highest class of similarity is determined as class belonging to the day image data of prediction.
Optionally, the step of clustering to history day image data includes: using two points of k means clustering algorithms, spectral clusterings
Any one clustering algorithm in algorithm, SOM clustering algorithm and FCM clustering algorithm clusters historical weather data.
Optionally, astronomical phenomena data include at least one of the following: the solar flare grade, the height of the moon, the orientation of the moon of the sun
Angle, the right ascension of the moon, the declination of the moon, the speed of service of the moon, the distance of the moon to the earth, at least one planet brightness,
The height of at least one planet, the right ascension of at least one planet, the declination of at least one planet, at least one planet to the sun
Distance.
Optionally, prediction technique further include: before establishing generated power forecasting model, history day image data is returned
One change processing.
Optionally, the step of history day image data being normalized includes: using linear function normalization algorithm
Or history day image data is normalized in standard deviation standardized algorithm.
Another aspect of the present invention provides a kind of pre- measurement equipment of the generated output of wind power generating set, pre- measurement equipment packet
Include: model foundation unit is sent out according to the history of the history day image data of wind power generating set position and wind power generating set
Electrical power establishes generated power forecasting model, wherein generated power forecasting model indicates between day image data and generated output
Corresponding relationship；Predicting unit, according to the day image data of prediction, by using generated power forecasting model prediction wind power generating set
Generated output.
Optionally, pre- measurement equipment further include: cluster cell, model foundation unit establish generated power forecasting model it
Before, history day image data is clustered, history day image data is divided into multiple classes, wherein generated power forecasting model
Including multiple neural network models, each neural network model corresponds to a class in multiple classes, wherein model foundation unit
Neural network model corresponding with each class is established according to the history day image data of each class and history generated output.
Optionally, predicting unit determines class belonging to the day image data of prediction, according to the day image data of prediction, by using
Neural network model corresponding with class belonging to the day image data of prediction, predicts the generated output of wind power generating set.
Optionally, predicting unit determines the similarity of the cluster centre of each class and the day image data of prediction, and will be similar
It spends highest class and is determined as class belonging to the day image data of prediction.
Optionally, cluster cell is using two points of k means clustering algorithms, spectral clustering, SOM clustering algorithm and FCM cluster
Any one clustering algorithm in algorithm clusters historical weather data.
Optionally, astronomical phenomena data include at least one of the following: the solar flare grade, the height of the moon, the orientation of the moon of the sun
Angle, the right ascension of the moon, the declination of the moon, the speed of service of the moon, the distance of the moon to the earth, at least one planet brightness,
The height of at least one planet, the right ascension of at least one planet, the declination of at least one planet, at least one planet to the sun
Distance.
Optionally, pre- measurement equipment further include: processing unit, model foundation unit establish generated power forecasting model it
Before, history day image data is normalized.
Optionally, processing unit is using linear function normalization algorithm or standard deviation standardized algorithm to history day image data
It is normalized.
Another aspect of the present invention provides a kind of computer readable storage medium, which has
Processor is made to execute the computer of the prediction technique of the generated output of wind power generating set as above when being executed by a processor
Program.
Another aspect of the present invention provides a kind of computing device, which includes: processor；Memory, for depositing
Storage, which is worked as, to be executed by processor so that processor executes the computer of the prediction technique of the generated output of wind power generating set as above
Program.
The prediction technique of the generated output of the wind power generating set of embodiment according to the present invention, for existing prediction side
The deficiency of method, finding and causing the principal element of generated power forecasting inaccuracy is existing data (weather data) sheet for predicting dependence
The accuracy of body is not high, and the input data of prediction model is replaced with the higher day image data of accuracy, improves generated output
The accuracy of prediction.
In addition, in order to further increase the accuracy of generated power forecasting, the wind-power electricity generation of embodiment according to the present invention
The prediction technique of the generated output of unit also clusters history day image data, then establishes power generation for different classes respectively
Power prediction model, since different generated power forecasting models consider the data characteristic of different classes, forecasting accuracy is higher.
Detailed description of the invention
By the detailed description carried out below in conjunction with the accompanying drawings, above and other objects of the present invention, features and advantages will
It becomes more fully apparent, in which:
Fig. 1 is the process for showing the prediction technique of the generated output of wind power generating set of embodiment according to the present invention
Figure；
Fig. 2 is the block diagram of the pre- measurement equipment of the generated output for the wind power generating set for showing embodiment according to the present invention.
Specific embodiment
Detailed description of the present invention embodiment with reference to the accompanying drawings.
Fig. 1 is the process for showing the prediction technique of the generated output of wind power generating set of embodiment according to the present invention
Figure.
In step S10, according to the history day image data of wind power generating set position and the history of wind power generating set
Generated output establishes generated power forecasting model.Generated power forecasting model indicates corresponding between day image data and generated output
Relationship.
The history day image data of wind power generating set position can be obtained from the Astronomical Observations Related mechanism of profession.For example, can
It first determines the longitude and latitude of wind power generating set position, then obtains the astronomical phenomena from the longitude and latitude from the Astronomical Observations Related mechanism of profession
Data.
As an example, day image data may include at least one of following: solar flare grade, the height of the moon, the moon of the sun
Azimuth, the right ascension of the moon, the declination of the moon, the speed of service of the moon, the distance of the moon to the earth, at least one planet it is bright
Degree, the declination of the height of at least one planet, the right ascension of at least one planet, at least one planet, at least one planet are to too
The distance of sun.Here planet can be the eight major planets of the solar system, comprising: Mercury, Venus, the earth, Mars, Jupiter, Saturn, Uranus and
Neptune.
In a preferred embodiment, in order to keep subsequent data processing more convenient, generated output is being established
Before prediction model, history day image data can be also normalized.
Here, various normalization algorithms can be used carry out history astronomical phenomena data and be normalized.For example, can be used linear
History day image data is normalized in normalization algorithm or standard deviation standardized algorithm.
As an example, following formula (1) can be used history day image data to be normalized.
Xnorm=(X-Xmin)/(Xmax–Xmin) (1)
Wherein, XnormIt is the history day image data after normalization, XmaxAnd XminRespectively original history day image data
In maximum value and minimum value.
It is appreciated that the present invention does not limit the specific algorithm of normalized, other normalization also can be used
Algorithm is normalized history day image data.
It in another preferred embodiment, can in order to further increase the forecasting accuracy of generated power forecasting model
Before establishing generated power forecasting model, history day image data is clustered, history day image data is drawn
It is divided into multiple classes.In the case where having carried out normalized to history day image data, to the history astronomical phenomena after normalized
Data are clustered.The detailed process clustered to history day image data will describe in detail below.
In this preferred embodiment, the generated power forecasting model include multiple neural network models (such as: it is reversed
Propagation Neural Network model, genetic algorithm back propagation neural network model etc.), each neural network model corresponds to a class in multiple classes.
Neural network corresponding with each class is established according to the history day image data of each class and history generated output
Model.That is, will be gone through using the history day image data of each class as the input of neural network model corresponding with each class
Output of the history generated output as neural network model corresponding with each class, to establish neural network mould corresponding with each class
Type.Here history generated output refers within the period corresponding with the history day image data of each class, wind power generating set
Generated output.Neural network model corresponding with each class indicates the corresponding pass of day image data and generated output in each class
System.
It is appreciated that in the case where not clustered to history day image data, it can be according to all history day image datas
And history generated output establishes generated power forecasting model.That is, using all history day image datas as power generation
The input of power prediction model, using history generated output as the output of generated power forecasting model, Lai Jianli generated output is pre-
Survey model.
It will be described in detail the detailed process clustered to history day image data below.
Here, various clustering algorithms can be used to cluster historical weather data.For example, it is poly- that two points of k mean values can be used
Class algorithm, spectral clustering, Self-organizing Maps (Self-Organizing Maps, SOM) clustering algorithm and fuzzy C-mean algorithm
Any one clustering algorithm in (Fuzzy C-means, FCM) clustering algorithm clusters historical weather data.It can manage
Solution, the present invention do not limit the specific algorithm of cluster, can also be carried out using other algorithms to historical weather data
Cluster.
Illustrate the process clustered to history day image data by taking two points of k means clustering algorithms as an example below.
First using history day image data as a class, such is then divided into two classes.
K- mean cluster is carried out to each class again, i.e., to each class, calculates cost function, i.e. square error and (Sum of
Squared errors, SSE), SSE value is smaller to indicate distance center of the data point closer to them, and Clustering Effect is also better,
Therefore selection is divided into two classes, until the number of class is equal to so that the smallest that class of SSE carries out division operation next time
Until given class number.
Shown in the calculation method of the SSE of each class such as following formula (2).
Wherein, CjFor the cluster centre of each class, XiFor the data point in each class, i indicates the number of data point, and m is indicated
The quantity for the data point that each class includes.
In step S20, according to the day image data of prediction, by using generated power forecasting model prediction wind power generating set
Generated output.That is, according to the day image data in the following predetermined amount of time of prediction, by using generated power forecasting
Generated output of the wind power generating set in the following predetermined amount of time in model prediction.Particularly, by the pre- timing of the future of prediction
Between day image data in section be input in generated power forecasting model, generated power forecasting model exports wind power generating set not
Carry out the generated output of predetermined amount of time.
In the case where generated power forecasting model includes multiple neural network models, the day image data of prediction can be first determined
Affiliated class, further according to the day image data of prediction, by using the corresponding nerve of class belonging to the day image data with the prediction
Network model predicts the generated output of wind power generating set.That is, by the astronomical phenomena number in the following predetermined amount of time of prediction
According to being input in above-mentioned corresponding neural network model, the corresponding neural network model output wind power generating set is following pre-
The generated output for section of fixing time.
It here, can be by the similarity of the day image data of the cluster centre and prediction of each class of determination, to determine prediction
Class belonging to its image data.Particularly, the highest class of similarity is determined as to class belonging to the day image data of prediction.Generally
Come, in cluster field, similarity is described with distance, that is to say, that with the day image data of prediction and the cluster centre of each class
The distance between, indicate the similarity of the cluster centre of each class and the day image data of prediction.Various existing calculating can be used
The method of the distance between sample calculates the day image data of prediction and the distance between the cluster centre of each class, for example, Europe
Family name's Furthest Neighbor, manhatton distance method, included angle cosine Furthest Neighbor or angle distance method etc..
Fig. 2 is the block diagram of the pre- measurement equipment of the generated output for the wind power generating set for showing embodiment according to the present invention.
The pre- measurement equipment of the generated output of the wind power generating set of embodiment according to the present invention includes model foundation unit 10 and prediction
Unit 20.
Model foundation unit 10 is according to the history day image data of wind power generating set position and wind power generating set
History generated output establishes generated power forecasting model.Generated power forecasting model indicates between day image data and generated output
Corresponding relationship.
The history day image data of wind power generating set position can be obtained from the Astronomical Observations Related mechanism of profession.For example, can
It first determines the longitude and latitude of wind power generating set position, then obtains the astronomical phenomena from the longitude and latitude from the Astronomical Observations Related mechanism of profession
Data.
As an example, day image data may include at least one of following: solar flare grade, the height of the moon, the moon of the sun
Azimuth, the right ascension of the moon, the declination of the moon, the speed of service of the moon, the distance of the moon to the earth, at least one planet it is bright
Degree, the declination of the height of at least one planet, the right ascension of at least one planet, at least one planet, at least one planet are to too
The distance of sun.Here planet can be the eight major planets of the solar system, comprising: Mercury, Venus, the earth, Mars, Jupiter, Saturn, Uranus and
Neptune.
In a preferred embodiment, in order to keep subsequent data processing more convenient, implement according to the present invention
The pre- measurement equipment of the generated output of the wind power generating set of example may also include processing unit (not shown).In model foundation list
Member 10 is established before generated power forecasting model, and history day image data can be normalized in processing unit.
Here, various normalization algorithms can be used carry out history astronomical phenomena data and be normalized.For example, can be used linear
History day image data is normalized in normalization algorithm or standard deviation standardized algorithm.
As an example, formula as described above (1) can be used history day image data to be normalized.
It is appreciated that the present invention does not limit the specific algorithm of normalized, other normalization also can be used
Algorithm is normalized history day image data.
In another preferred embodiment, in order to improve the forecasting accuracy of generated power forecasting model, according to this hair
The pre- measurement equipment of the generated output of the wind power generating set of bright embodiment may also include cluster cell (not shown).In model
Unit 10 is established before establishing generated power forecasting model, cluster cell clusters history day image data, will be described
History day image data is divided into multiple classes.In the case where having carried out normalized to history day image data, at normalization
History day image data after reason is clustered.The detailed process that cluster cell clusters history day image data can refer to above
The detailed process clustered in text to history day image data, details are not described herein.
In this preferred embodiment, the generated power forecasting model includes that multiple neural network models are (such as reversed
Propagation Neural Network model, genetic algorithm back propagation neural network model etc.), each neural network model corresponds to a class in multiple classes.
Model foundation unit 10 is established and each class according to the history day image data of each class and history generated output
Corresponding neural network model.That is, using the history day image data of each class as neural network corresponding with each class
The input of model, using history generated output as the output of neural network model corresponding with each class, to establish and each class
Corresponding neural network model.Here history generated output referred in the period corresponding with the history day image data of each class
It is interior, the generated output of wind power generating set.Neural network model corresponding with each class indicate day image data in each class with
The corresponding relationship of generated output.
It is appreciated that in the case where not clustered to history day image data, it can be according to all history day image datas
And history generated output establishes generated power forecasting model.That is, using all history day image datas as power generation
The input of power prediction model, using history generated output as the output of generated power forecasting model, Lai Jianli generated output is pre-
Survey model.
Predicting unit 20 is according to the day image data of prediction, by using generated power forecasting model prediction wind power generating set
Generated output.That is, according to the day image data in the following predetermined amount of time of prediction, by using generated power forecasting
Generated output of the wind power generating set in the following predetermined amount of time in model prediction.Particularly, by the pre- timing of the future of prediction
Between day image data in section be input in generated power forecasting model, generated power forecasting model exports wind power generating set not
Carry out the generated output of predetermined amount of time.
In the case where generated power forecasting model includes multiple neural network models, predicting unit 20 can first determine prediction
Day image data belonging to class, further according to the day image data of prediction, by using class belonging to the day image data with the prediction
Corresponding neural network model predicts the generated output of wind power generating set.That is, by the following predetermined amount of time of prediction
Interior day image data is input in above-mentioned corresponding neural network model, which exports wind-driven generator
Generated output of the group in the following predetermined amount of time.
Here, predicting unit 20 can be come by the similarity of the day image data of the cluster centre and prediction of each class of determination
Determine class belonging to the day image data of prediction.Particularly, belonging to the day image data that the highest class of similarity is determined as to prediction
Class.It is, in general, that in cluster field, similarity is described with distance, that is to say, that day image data and each class with prediction
The distance between cluster centre, indicate the similarity of the cluster centre of each class and the day image data of prediction.It can be used various
The existing method for calculating the distance between sample calculate between the day image data of prediction and the cluster centre of each class away from
From for example, Euclidean distance method, manhatton distance method, included angle cosine Furthest Neighbor or angle distance method etc..
The prediction technique and equipment of the generated output of the wind power generating set of embodiment according to the present invention, for existing
The deficiency of prediction technique, finding and causing the principal element of generated power forecasting inaccuracy is the existing data (weather for predicting to rely on
Data) itself accuracy it is not high, the input data of prediction model is replaced with into the higher day image data of accuracy, improves hair
The accuracy of electrical power prediction.
In addition, in order to further increase the accuracy of generated power forecasting, the wind-power electricity generation of embodiment according to the present invention
The prediction technique and equipment of the generated output of unit also cluster history day image data, then build respectively for different classes
Vertical generated power forecasting model, since different generated power forecasting models consider the data characteristic of different classes, prediction is accurate
Property is higher.
Embodiment according to the present invention also provides a kind of computer readable storage medium.The computer readable storage medium is deposited
Contain the prediction technique for the generated output for making processor execute wind power generating set as described above when being executed by a processor
Computer program.
Embodiment according to the present invention also provides a kind of computing device.The computing device includes processor and memory.It deposits
Reservoir is for storing program instruction.Described program instruction is executed by processor so that processor executes wind-power electricity generation as described above
The computer program of the prediction technique of the generated output of unit.
In addition, each program in the pre- measurement equipment of the generated output of the wind power generating set of embodiment according to the present invention
Module can be realized by hardware completely, such as field programmable gate array or specific integrated circuit；It can also be by hardware and software
The mode that combines is realized；It can also be realized completely by computer program with software mode.
Although being particularly shown and describing the present invention, those skilled in the art referring to its exemplary embodiment
It should be understood that in the case where not departing from the spirit and scope of the present invention defined by claim form can be carried out to it
With the various changes in details.
Claims (14)
1. a kind of prediction technique of the generated output of wind power generating set, which is characterized in that the prediction technique includes:
According to the history of the history day image data of the wind power generating set position and wind power generating set power generation function
Rate establishes generated power forecasting model, wherein between the generated power forecasting model instruction day image data and generated output
Corresponding relationship；
According to the day image data of prediction, by using the power generation of wind power generating set described in the generated power forecasting model prediction
Power.
2. prediction technique according to claim 1, which is characterized in that the prediction technique further include:
Before establishing the generated power forecasting model, history day image data is clustered, by the history day
Image data is divided into multiple classes,
Wherein, the generated power forecasting model includes multiple neural network models, and each neural network model corresponds to described
A class in multiple classes,
Wherein, the step of establishing generated power forecasting model includes: the history day image data and the history according to each class
Generated output establishes neural network model corresponding with each class.
3. prediction technique according to claim 2, which is characterized in that predict the generated output of the wind power generating set
Step comprises determining that class belonging to the day image data of the prediction, according to the day image data of the prediction, by using with it is described
The corresponding neural network model of class, predicts the generated output of the wind power generating set belonging to the day image data of prediction.
4. prediction technique according to claim 3, which is characterized in that determine class belonging to the day image data of the prediction
Step comprises determining that the similarity of the cluster centre of each class and the day image data of the prediction, and by the highest class of similarity
It is determined as class belonging to the day image data of the prediction.
5. prediction technique according to claim 1, which is characterized in that astronomical phenomena data include at least one of the following: the sun
Solar flare grade, the height of the moon, the azimuth of the moon, the right ascension of the moon, the declination of the moon, the speed of service of the moon, the moon are extremely
The distance of the earth, the brightness of at least one planet, the height of at least one planet, at least one planet right ascension, at least one
The distance of the declination of planet, at least one planet to the sun.
6. prediction technique according to claim 1, which is characterized in that the prediction technique further include:
Before establishing the generated power forecasting model, history day image data is normalized.
7. a kind of pre- measurement equipment of the generated output of wind power generating set, which is characterized in that the pre- measurement equipment includes:
Model foundation unit, according to the history day image data of the wind power generating set position and the wind power generating set
History generated output establish generated power forecasting model, wherein generated power forecasting model instruction day image data and hair
Corresponding relationship between electrical power；
Predicting unit, according to the day image data of prediction, by using wind-power electricity generation described in the generated power forecasting model prediction
The generated output of unit.
8. pre- measurement equipment according to claim 7, which is characterized in that the pre- measurement equipment further include:
Cluster cell, before model foundation unit establishes the generated power forecasting model, to history day image data into
Row cluster, is divided into multiple classes for history day image data,
Wherein, the generated power forecasting model includes multiple neural network models, and each neural network model corresponds to described
A class in multiple classes,
Wherein, the model foundation unit according to the history day image data of each class and the history generated output establish with it is every
The corresponding neural network model of a class.
9. pre- measurement equipment according to claim 8, which is characterized in that the predicting unit determines the astronomical phenomena number of the prediction
It is corresponding by using class belonging to the day image data with the prediction according to the day image data of the prediction according to affiliated class
Neural network model predicts the generated output of the wind power generating set.
10. pre- measurement equipment according to claim 9, which is characterized in that the predicting unit determines in the cluster of each class
The similarity of the heart and the day image data of the prediction, and the highest class of similarity is determined as belonging to the day image data of the prediction
Class.
11. pre- measurement equipment according to claim 7, which is characterized in that astronomical phenomena data include at least one of the following: the sun
Solar flare grade, the height of the moon, the azimuth of the moon, the right ascension of the moon, the declination of the moon, the speed of service of the moon, the moon are extremely
The distance of the earth, the brightness of at least one planet, the height of at least one planet, at least one planet right ascension, at least one
The distance of the declination of planet, at least one planet to the sun.
12. pre- measurement equipment according to claim 7, which is characterized in that the pre- measurement equipment further include: processing unit,
Model foundation unit is established before the generated power forecasting model, and history day image data is normalized.
13. a kind of computer readable storage medium is stored with and processor is made to execute such as claim 1 when being executed by a processor
To the computer program of the prediction technique of the generated output of wind power generating set described in any one of 6.
14. a kind of computing device, comprising:
Processor；
Memory is executed by processor for storing to work as so that processor is executed as described in any one of claim 1 to 6
The computer program of the prediction technique of the generated output of wind power generating set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810210270.3A CN110276468A (en) | 2018-03-14 | 2018-03-14 | Method and device for predicting generated power of wind generating set |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810210270.3A CN110276468A (en) | 2018-03-14 | 2018-03-14 | Method and device for predicting generated power of wind generating set |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110276468A true CN110276468A (en) | 2019-09-24 |
Family
ID=67957789
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810210270.3A Pending CN110276468A (en) | 2018-03-14 | 2018-03-14 | Method and device for predicting generated power of wind generating set |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110276468A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102184453A (en) * | 2011-05-16 | 2011-09-14 | 上海电气集团股份有限公司 | Wind power combination predicting method based on fuzzy neural network and support vector machine |
| CN104299044A (en) * | 2014-07-01 | 2015-01-21 | 沈阳工程学院 | Clustering-analysis-based wind power short-term prediction system and prediction method |
| CN104933483A (en) * | 2015-06-17 | 2015-09-23 | 中国电力科学研究院 | Wind power forecasting method dividing based on weather process |
| CN105303268A (en) * | 2015-11-25 | 2016-02-03 | 国家电网公司 | Wind power generation output power prediction method based on similarity theory |
| CN105654207A (en) * | 2016-01-07 | 2016-06-08 | 国网辽宁省电力有限公司锦州供电公司 | Wind power prediction method based on wind speed information and wind direction information |
| CN105930900A (en) * | 2016-05-09 | 2016-09-07 | 华北电力大学 | Method and system for predicting hybrid wind power generation |
| CN107133695A (en) * | 2017-04-28 | 2017-09-05 | 北京天诚同创电气有限公司 | A kind of wind power forecasting method and system |
| CN107609697A (en) * | 2017-09-06 | 2018-01-19 | 南京邮电大学 | A kind of Wind power forecasting method |
-
2018
- 2018-03-14 CN CN201810210270.3A patent/CN110276468A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102184453A (en) * | 2011-05-16 | 2011-09-14 | 上海电气集团股份有限公司 | Wind power combination predicting method based on fuzzy neural network and support vector machine |
| CN104299044A (en) * | 2014-07-01 | 2015-01-21 | 沈阳工程学院 | Clustering-analysis-based wind power short-term prediction system and prediction method |
| CN104933483A (en) * | 2015-06-17 | 2015-09-23 | 中国电力科学研究院 | Wind power forecasting method dividing based on weather process |
| CN105303268A (en) * | 2015-11-25 | 2016-02-03 | 国家电网公司 | Wind power generation output power prediction method based on similarity theory |
| CN105654207A (en) * | 2016-01-07 | 2016-06-08 | 国网辽宁省电力有限公司锦州供电公司 | Wind power prediction method based on wind speed information and wind direction information |
| CN105930900A (en) * | 2016-05-09 | 2016-09-07 | 华北电力大学 | Method and system for predicting hybrid wind power generation |
| CN107133695A (en) * | 2017-04-28 | 2017-09-05 | 北京天诚同创电气有限公司 | A kind of wind power forecasting method and system |
| CN107609697A (en) * | 2017-09-06 | 2018-01-19 | 南京邮电大学 | A kind of Wind power forecasting method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Pearre et al. | Statistical approach for improved wind speed forecasting for wind power production | |
| CN113128793A (en) | Photovoltaic power combination prediction method and system based on multi-source data fusion | |
| CN103390116B (en) | Use the photovoltaic power station power generation power forecasting method of stepping mode | |
| Stefenon et al. | Photovoltaic power forecasting using wavelet Neuro-Fuzzy for active solar trackers | |
| CN107909211B (en) | Wind field equivalent modeling and optimization control method based on fuzzy c-means clustering algorithm | |
| Melzi et al. | Hourly solar irradiance forecasting based on machine learning models | |
| CN103942736B (en) | A kind of wind power plant multimachine equivalent modeling method | |
| CN108196317B (en) | Meteorological prediction method for micro-grid system | |
| CN107749638B (en) | Multi-microgrid combined virtual power plant distributed random non-overlapping sampling centerless optimization method | |
| CN112994092B (en) | Independent wind-solar storage micro-grid system size planning method based on power prediction | |
| CN114021420A (en) | Distributed photovoltaic ultra-short-term power generation prediction method and system | |
| CN114792156A (en) | Photovoltaic output power prediction method and system based on curve characteristic index clustering | |
| CN112070311A (en) | Day-ahead light power prediction method based on similar day clustering and meteorological factor weighting | |
| Park et al. | Multi-layer RNN-based short-term photovoltaic power forecasting using IoT dataset | |
| CN115564108A (en) | Deep peak regulation oriented optimal scheduling method for light storage and load in virtual power plant | |
| CN116306006A (en) | Optimal scheduling method and device for wind-solar-water combined power generation and computer equipment | |
| Das et al. | Optimized support vector regression-based model for solar power generation forecasting on the basis of online weather reports | |
| CN109767353A (en) | A kind of photovoltaic power generation power prediction method based on probability-distribution function | |
| Lee et al. | Bayesian deep learning-based confidence-aware solar irradiance forecasting system | |
| CN109586309B (en) | Power distribution network reactive power optimization method based on big data free entropy theory and scene matching | |
| CN116454872A (en) | Wind power prediction method and device, electronic equipment and storage medium | |
| CN116484998A (en) | Distributed photovoltaic power station power prediction method and system based on meteorological similar day | |
| CN110276468A (en) | Method and device for predicting generated power of wind generating set | |
| CN105205560A (en) | Photovoltaic power supply power prediction method based on positive and negative error variable weights | |
| CN113991711A (en) | Capacity configuration method for energy storage system of photovoltaic power station |
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 |
Application publication date: 20190924 |
|
| RJ01 | Rejection of invention patent application after publication |