CN107194499A - The Forecasting Methodology and prediction meanss of region wind-powered electricity generation short term power - Google Patents
The Forecasting Methodology and prediction meanss of region wind-powered electricity generation short term power Download PDFInfo
- Publication number
- CN107194499A CN107194499A CN201710295885.6A CN201710295885A CN107194499A CN 107194499 A CN107194499 A CN 107194499A CN 201710295885 A CN201710295885 A CN 201710295885A CN 107194499 A CN107194499 A CN 107194499A
- Authority
- CN
- China
- Prior art keywords
- wind power
- power plant
- region
- wind
- matrix
- 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
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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- 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
Abstract
The present invention relates to a kind of Forecasting Methodology of region wind-powered electricity generation short term power, wherein, methods described includes：Wind power plant in region is clustered, and is many sub-regions by region division according to cluster result；Obtain the relative coefficient exerted oneself with region wind power output of each wind power plant in region；Obtain the predicated error of each wind power plant in each subregion；In every sub-regions, a wind power plant is selected as representing wind power plant according to relative coefficient and predicated error；The weight coefficient for representing wind power plant is calculated according to the installed capacity of subregion；And wind farm power prediction value and weight coefficient are represented according to each, obtain region wind power prediction value.The invention further relates to a kind of prediction meanss of region wind-powered electricity generation short term power.Above-mentioned Forecasting Methodology and prediction meanss, can reduce dependence of the regional prediction model to single wind farm data completeness and precision of prediction.
Description
Technical field
The present invention relates to Forecasting Methodology and the prediction of field of new energy generation, more particularly to a kind of region wind-powered electricity generation short term power
Device.
Background technology
Wind power prediction is the key foundation of new energy scheduling, and unit group a few days ago is optimized according to wind power output prediction curve
Close, dynamic rolling adjustment conventional power unit is exerted oneself, and spare capacity is reduced, so as to reduce system operation cost.Wind power prediction is led to
Often carry out for single wind power plant, but the power supply for being enriched with area with wind-resources is concentrated and developed, and has gradually formed large-scale wind power
Base, the prediction of region wind power output total value becomes more and more important.
Traditional regional wind power prediction method is single game power summation, i.e., each wind power in region is pre-
Survey result and obtain region wind power prediction value with markers addition.But there is obvious limitation in this method, such as：1. for number
According to the incomplete wind power plant of condition, indivedual wind farm power prediction low precisions may tie down whole region precision of prediction；2. it is newly-built
Wind power plant not yet sets up forecasting system at grid-connected initial stage；3. it is not suitable for having the more wind power prediction functional module that do not configure
The region of distributing wind power integration.
The content of the invention
In summary, in view of the above-mentioned problems, it is necessory to provide the Forecasting Methodology of region wind-powered electricity generation short term power a kind of and pre-
Survey device.
A kind of Forecasting Methodology of region wind-powered electricity generation short term power, wherein, methods described includes：
Wind power plant in region is clustered, and is many sub-regions by region division according to cluster result；
Obtain the relative coefficient exerted oneself with region wind power output of each wind power plant in region；
Obtain the predicated error of each wind power plant in each subregion；
In every sub-regions, at least one wind power plant is selected as representing wind-powered electricity generation according to relative coefficient and predicated error
；
The weight coefficient for representing wind power plant is calculated according to the installed capacity of subregion；And
Wind farm power prediction value and weight coefficient are represented according to each, region wind power prediction value is obtained.
The Forecasting Methodology of the region wind-powered electricity generation short term power provided relative to conventional art, the present invention, by obtaining in region
Wind power plant and the mode of weight coefficient are represented, regional prediction model can be reduced to single wind farm data completeness and precision of prediction
Dependence.
As one of embodiment, the wind power plant in region is clustered, and according to cluster result by region
The step of being divided into many sub-regions includes：
Historical power sequence to wind power plant in region carries out empirical orthogonal function decomposition, obtains spatial mode matrix；
The spatial mode matrix obtained using hierarchical clustering method to empirical orthogonal function decomposition carries out clustering, is clustered
As a result；And
It is some subregions by region division based on hierarchical clustering result.
As one of embodiment, the historical power sequence to wind power plant in region carries out Empirical Orthogonal Function point
Solution, the step of obtaining spatial mode matrix includes：
The historical power data sample of all wind power plants in region is obtained, raw data matrix X is obtained；
Raw data matrix X is standardized, normalized matrix Y is obtained；
Normalized matrix Y covariance matrix R, wherein；
Calculating matrix R characteristic root, and, obtain the characteristic vector of each characteristic root by characteristic root in magnitude order, as
Each spatial mode of original wind power plant, forms spatial mode matrix.
As one of embodiment, the spatial mode square that the use hierarchical clustering method is obtained to empirical orthogonal function decomposition
Battle array carries out clustering, and the step of obtaining cluster result includes：
The N number of class of initial construction, N is the quantity of wind power plant to be divided here；
The Euclidean distance of N number of sample between any two is calculated, distance matrix d is obtained；
Between class distance is calculated, it is determined that simultaneously combined distance most close 2 classes are 1 new class；
Whether the quantity for judging the class after cluster is 1, if the number of class is not equal to 1, computes repeatedly between class distance, it is determined that
And combined distance most close 2 classes are 1 new class, until the quantity of the class after cluster is 1；
Obtain hierarchical clustering result.
It is described that at least one wind power plant conduct is selected according to relative coefficient and predicated error as one of embodiment
The step of representing wind power plant includes：
In every sub-regions, one relative coefficient of selection and prediction level are all higher than the wind power plant conduct of average level
Represent wind power plant.
A kind of prediction meanss of region wind-powered electricity generation short term power, wherein, described device includes：
Region division module, be by region division for being clustered to the wind power plant in region, and according to cluster result
Many sub-regions；
Relative coefficient acquisition module, the correlation exerted oneself with region wind power output for obtaining each wind power plant in region
Coefficient；
Error acquisition module, the predicated error for obtaining each wind power plant in each subregion；
Wind power plant selecting module, in every sub-regions, according to relative coefficient and predicated error selection at least one
Individual wind power plant is as representing wind power plant；
Weight coefficient computing module, the weight coefficient of wind power plant is represented for being calculated according to the installed capacity of subregion；
Wind power prediction module, for representing wind farm power prediction value and weight coefficient, zoning according to each
Wind power prediction value.
The prediction meanss of the region wind-powered electricity generation short term power provided relative to conventional art, the present invention, by determining in region
Wind power plant and the mode of weight coefficient are represented, regional prediction model can be reduced to single wind farm data completeness and precision of prediction
Dependence.
As one of embodiment, the region division module includes：
Spatial mode matrix acquiring unit, Empirical Orthogonal Function point is carried out for the historical power sequence to wind power plant in region
Solution, obtains spatial mode matrix；
Cluster cell, the spatial mode matrix for being obtained using hierarchical clustering method to empirical orthogonal function decomposition is clustered
Analysis, obtains hierarchical clustering result；
Sub-zone dividing unit, for being some subregions by region division based on hierarchical clustering result.
As one of embodiment, the spatial mode matrix acquiring unit is additionally operable to：
The historical power data sample of all wind power plants in region is obtained, raw data matrix X is obtained；
Raw data matrix X is standardized, normalized matrix Y is obtained；
Normalized matrix Y covariance matrix R, wherein；
Calculating matrix R characteristic root, and, obtain the characteristic vector of each characteristic root by characteristic root in magnitude order, as
Each spatial mode of original wind power plant, forms spatial mode matrix.
As one of embodiment, cluster cell is additionally operable to：
The N number of class of initial construction, N is the quantity of wind power plant to be divided here；
The Euclidean distance of N number of sample between any two is calculated, distance matrix d is obtained；
Between class distance is calculated, it is determined that simultaneously combined distance most close 2 classes are 1 new class；
Whether the quantity for judging the class after cluster is 1, if the number of class is not equal to 1, between class distance is calculated again, it is determined that
And combined distance most close 2 classes are 1 new class, until the quantity of the class after cluster is 1；
Obtain hierarchical clustering result.
As one of embodiment, the wind power plant selecting module is additionally operable to：
In every sub-regions, one relative coefficient of selection and prediction level are all higher than the wind power plant conduct of average level
Represent wind power plant.
Brief description of the drawings
Fig. 1 is the FB(flow block) of the Forecasting Methodology of wind-powered electricity generation short term power in region provided in an embodiment of the present invention；
Fig. 2 is the structured flowchart of the prediction meanss of wind-powered electricity generation short term power in region provided in an embodiment of the present invention.
Embodiment
Further stated in detail below according to Figure of description and in conjunction with specific embodiments to technical scheme.
Rise scale prediction refer to set up region power generating value and some wind power plants with available online data power generating value it
Between Function Mapping relation, can accurately be predicted whole under conditions of only there is a number of wind farm data
The wind power output in region.Therefore, the present invention proposes that a kind of region wind-powered electricity generation short term power statistics rises scale prediction method, so as to reduce
Dependence of the regional prediction model to single wind farm data completeness and precision of prediction.
Referring to Fig. 1, the Forecasting Methodology of wind-powered electricity generation short term power in region provided in an embodiment of the present invention, comprises the following steps：
Step S10, is clustered to the wind power plant in region, and is multiple sub-districts by region division according to cluster result
Domain；
Step S20, obtains the relative coefficient exerted oneself with region wind power output of each wind power plant in region；
Step S30, obtains the predicated error of each wind power plant in each subregion；
Step S40, in every sub-regions, at least one wind power plant conduct is selected according to relative coefficient and predicated error
Represent wind power plant；
Step S50, the weight coefficient for representing wind power plant is calculated according to the installed capacity of subregion；
Step S60, wind farm power prediction value and weight coefficient, zoning wind power prediction are represented according to each
Value.
In step slo, the wind power plant in region is clustered, and is many height by region division according to cluster result
Region includes：
Step S11, the historical power sequence to wind power plant in region carries out Empirical Orthogonal Function (EOF) decomposition, obtains sky
Between type matrix.
Specifically, the historical power sequence to wind power plant in region carries out the detailed process bag of empirical orthogonal function decomposition
Include：
Step S111, obtains the historical power data sample of all wind power plants in region, obtains raw data matrix X；
Step S112, is standardized to raw data matrix X, obtains normalized matrix Y；
Step S113, normalized matrix Y covariance matrix R, wherein R=YYT；
Step S114, calculating matrix R characteristic root, and, obtain the feature of each characteristic root by characteristic root in magnitude order
Vector, as each spatial mode of original wind power plant, forms spatial mode matrix；
By characteristic root in magnitude order, λ is made1≥λ2≥...≥λM>=0, the corresponding characteristic vector of each characteristic root is respectively
V1,V2,...,VM, wherein V1,V2,...,VMEach spatial mode of as original wind power plant.
As one of embodiment, further, obtain after spatial mode matrix, may also include：
Step S115, calculates the variance contribution ratio of each spatial mode.
In step sl 15, spatial mode VmVariance contribution ratio beThe contribution rate of preceding k spatial mode
SumThe contribution rate of accumulative total of referred to as preceding k spatial mode.
Step S12, the spatial mode matrix obtained by clustering procedure to empirical orthogonal function decomposition carries out clustering, obtains
Cluster result.
The clustering procedure can be the methods such as hierarchical clustering method, K-means clustering procedures.Hierarchical clustering is used in the present embodiment
Method, the input data of hierarchical clustering method is that EOF decomposes the preceding k obtained higher spatial modes of accumulative variance contribution ratio, tool herein
Body step is as follows：
Step S121, the N number of class of initial construction, N is the quantity of wind power plant to be divided here；
Step S122, calculates the Euclidean distance of N number of sample between any two, obtains distance matrix d；
Step S123, calculates between class distance, it is determined that simultaneously combined distance most close 2 classes are 1 new class；
Step S124, whether the quantity for judging the class after cluster is 1, if the number of class is not equal to 1, repeats step
S123, until the quantity of the class after cluster is 1；
Step S125, obtains hierarchical clustering result.
In step S123, polymeric rule can be connect by sum of squares of deviations method, simply connected, is coupled completely between polymeric rule, class
Averagely connection polymeric rule, averagely couple in class in polymeric rule at least any one calculate between class distance.
Step S13, is some subregions by region division based on hierarchical clustering result.
According to hierarchical clustering result, the number of number subregion for needed for of class is taken, then in cluster result in same class
Wind power plant is divided into same sub-regions.
The step exerted oneself with the relative coefficient of region wind power output of each wind power plant in the acquisition region described in step S20
In rapid, each wind power historical data and region wind power historical data in calmodulin binding domain CaM obtain each wind in region
Relative coefficient between electric field and region wind power, such as following formula：
Wherein pkThe power sequence of k-th of wind power plant is represented,Represent the serial mean；pallRepresent region wind-powered electricity generation work(
Rate sequence,Represent the serial mean；N represents length of time series.
In step s 30, the predicated error includes the root-mean-square error and average absolute mistake of each wind-powered electricity generation field prediction power
Difference.
In step s 40, in every sub-regions, it can select a relative coefficient and prediction level all higher than average
The wind power plant of level is as representing wind power plant.
In step s 50, in per sub-regions, the weight coefficient for representing wind power plant can be by calculating all wind in such
The summation of electric field installed capacity represents the ratio of wind energy turbine set installed capacity with this.
In step S60, below equation meter can be passed through according to each power prediction value and its weight coefficient for representing wind power plant
Calculate region wind power prediction value：
pall,f=Σ bkpk,f
Wherein pall,fFor region wind power prediction value, pk,fTo represent the power prediction value of wind power plant, b for k-thkFor kth
The individual weight coefficient for representing wind power plant.
The Forecasting Methodology for the region wind-powered electricity generation short term power that the present invention is provided, by determining to represent wind power plant and weight in region
The mode of coefficient, is predicted without the pre- power scale to all wind power plants, it is only necessary to which the pre- power scale of part wind power plant is
The pre- power scale of region wind-powered electricity generation can be obtained, can reduce regional prediction model to single wind farm data completeness and precision of prediction according to
Rely.
Also referring to Fig. 2, the embodiment of the present invention further provides for a kind of prediction meanss of region wind-powered electricity generation short term power
1000, including：
Region division module 100, for being clustered to the wind power plant in region, and according to cluster result by region division
For many sub-regions；
Relative coefficient acquisition module 200, the phase exerted oneself with region wind power output for obtaining each wind power plant in region
Close property coefficient；
Error acquisition module 300, the predicated error for obtaining each wind power plant in each subregion；
Wind power plant selecting module 400, in every sub-regions, being selected at least according to relative coefficient and predicated error
One wind power plant is as representing wind power plant；
Weight coefficient computing module 500, the weight coefficient of wind power plant is represented for being calculated according to the installed capacity of subregion；
Wind power prediction module 600, for representing wind farm power prediction value and weight coefficient according to each, calculates area
Domain wind power prediction value.
The region division module 100 is used to cluster the wind power plant in region, and according to cluster result by region
Many sub-regions are divided into, the region division module 100 includes：
Spatial mode matrix acquiring unit 101, empirical orthogonal letter is carried out for the historical power sequence to wind power plant in region
Number (EOF) is decomposed, and obtains spatial mode matrix；
Cluster cell 103, the spatial mode matrix for being obtained using hierarchical clustering method to empirical orthogonal function decomposition is carried out
Clustering, obtains hierarchical clustering result；
Sub-zone dividing unit 105, for being some subregions by region division based on hierarchical clustering result.
Specifically, the spatial mode matrix acquiring unit 101 is additionally operable to：
The historical power data sample of all wind power plants in region is obtained, raw data matrix X is obtained；
Raw data matrix X is standardized, normalized matrix Y is obtained；
Normalized matrix Y covariance matrix R, wherein R=YYT；
Calculating matrix R characteristic root, and, obtain the characteristic vector of each characteristic root by characteristic root in magnitude order, as
Each spatial mode of primary field.
As one of embodiment, the spatial mode matrix acquiring unit 101 is additionally operable to：Calculate the variance tribute of each spatial mode
Offer rate.
The cluster cell 103 is additionally operable to：
The N number of class of initial construction, N is the quantity of wind power plant to be divided here；
The Euclidean distance of N number of sample between any two is calculated, distance matrix d is obtained；
Between class distance is calculated, it is determined that simultaneously combined distance most close 2 classes are 1 new class；
Whether the quantity for judging the class after cluster is 1, if the number of class is not equal to 1, reuses sum of squares of deviations method
Between class distance is calculated, it is determined that simultaneously combined distance most close 2 classes are 1 new class, until the quantity of the class after cluster is 1；
Obtain hierarchical clustering result.
The sub-zone dividing unit 105 is used for according to hierarchical clustering result, the number of the number taken subregion for needed for, will
Wind power plant in cluster result in same class is divided into same sub-regions.
Relative coefficient acquisition module 200 is used for each wind power historical data and region wind-powered electricity generation work(in calmodulin binding domain CaM
Rate historical data, obtains the relative coefficient between each wind power plant and region wind power, such as following formula：
Wherein pkIt is the power sequence of k-th of wind power plant,Represent the serial mean；pallIt is region wind power sequence
Row,Represent the serial mean；N represents length of time series.
The error acquisition module 300 can be used for obtaining r.m.s. mistake of the predicated error including each wind-powered electricity generation field prediction power
Difference and mean absolute error.
The wind power plant selecting module 400 can be used in every sub-regions, one relative coefficient of selection and prediction level
All higher than average level wind power plant as representing wind power plant.
The weight coefficient computing module 500 can be used for filling by every sub-regions, calculating all wind power plants in such
The ratio that the summation of machine capacity represents wind energy turbine set installed capacity with this is worth to the weight coefficient for representing wind power plant.
Wind power prediction module 600 can be used for according to each power prediction value and its weight coefficient for representing wind power plant, lead to
Cross below equation zoning wind power prediction value：
pall,f=∑ bkpk,f
Wherein pall,fFor region wind power prediction value, pk,fTo represent the power prediction value of wind power plant, b for k-thkFor kth
The individual weight coefficient for representing wind power plant.
The prediction meanss for the region wind-powered electricity generation short term power that the present invention is provided, by determining to represent wind power plant and weight in region
The mode of coefficient, is predicted without the pre- power scale to all wind power plants, it is only necessary to which the pre- power scale of part wind power plant is
The pre- power scale of region wind-powered electricity generation can be obtained, can reduce regional prediction model to single wind farm data completeness and precision of prediction according to
Rely.
For convenience of description, each several part of apparatus described above is divided into various modules with function or unit is described respectively.
Certainly, each module or the function of unit can be realized in same or multiple softwares or hardware when implementing the application.
It should be understood by those skilled in the art that, embodiments herein can be provided as method, system or computer program
Product.Therefore, the application can be using the reality in terms of complete hardware embodiment, complete software embodiment or combination software and hardware
Apply the form of example.Moreover, the application can be used in one or more computers for wherein including computer usable program code
The computer program production that usable storage medium is implemented on (including but is not limited to magnetic disk storage, CD-ROM, optical memory etc.)
The form of product.
The application is the flow with reference to method, equipment (system) and computer program product according to the embodiment of the present application
Figure and/or block diagram are described.It should be understood that can be by every first-class in computer program instructions implementation process figure and/or block diagram
Journey and/or the flow in square frame and flow chart and/or block diagram and/or the combination of square frame.These computer programs can be provided
The processor of all-purpose computer, special-purpose computer, Embedded Processor or other programmable data processing devices is instructed to produce
A raw machine so that produced by the instruction of computer or the computing device of other programmable data processing devices for real
The device for the function of being specified in present one flow of flow chart or one square frame of multiple flows and/or block diagram or multiple square frames.
These computer program instructions, which may be alternatively stored in, can guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works so that the instruction being stored in the computer-readable memory, which is produced, to be included referring to
Make the manufacture of device, the command device realize in one flow of flow chart or multiple flows and/or one square frame of block diagram or
The function of being specified in multiple square frames.
These computer program instructions can be also loaded into computer or other programmable data processing devices so that in meter
Series of operation steps is performed on calculation machine or other programmable devices to produce computer implemented processing, thus in computer or
The instruction performed on other programmable devices is provided for realizing in one flow of flow chart or multiple flows and/or block diagram one
The step of function of being specified in individual square frame or multiple square frames.
Embodiment described above only expresses the several embodiments of the present invention, and it describes more specific and detailed, but simultaneously
Therefore the limitation to the scope of the claims of the present invention can not be interpreted as.It should be pointed out that for one of ordinary skill in the art
For, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to the guarantor of the present invention
Protect scope.Therefore, the protection domain of patent of the present invention should be determined by the appended claims.
Claims (10)
1. a kind of Forecasting Methodology of region wind-powered electricity generation short term power, it is characterised in that methods described includes：
Wind power plant in region is clustered, and is many sub-regions by region division according to cluster result；
Obtain the relative coefficient exerted oneself with region wind power output of each wind power plant in region；
Obtain the predicated error of each wind power plant in each subregion；
In every sub-regions, at least one wind power plant is selected as representing wind power plant according to relative coefficient and predicated error；
The weight coefficient for representing wind power plant is calculated according to the installed capacity of subregion；And
Wind farm power prediction value and weight coefficient are represented according to each, region wind power prediction value is obtained.
2. according to the method described in claim 1, it is characterised in that the wind power plant in region is clustered, and according to
The step of region division is many sub-regions by cluster result includes：
Historical power sequence to wind power plant in region carries out empirical orthogonal function decomposition, obtains spatial mode matrix；
The spatial mode matrix obtained using hierarchical clustering method to empirical orthogonal function decomposition carries out clustering, obtains cluster knot
Really；And
It is some subregions by region division based on hierarchical clustering result.
3. method according to claim 2, it is characterised in that the historical power sequence to wind power plant in region is carried out
Empirical orthogonal function decomposition, the step of obtaining spatial mode matrix includes：
The historical power data sample of all wind power plants in region is obtained, raw data matrix X is obtained；
Raw data matrix X is standardized, normalized matrix Y is obtained；
Normalized matrix Y covariance matrix R, wherein R=YYT；
Calculating matrix R characteristic root, and, the characteristic vector of each characteristic root is obtained, as original by characteristic root in magnitude order
Each spatial mode of wind power plant, forms spatial mode matrix.
4. method according to claim 2, it is characterised in that the use hierarchical clustering method is to empirical orthogonal function decomposition
Obtained spatial mode matrix carries out clustering, and the step of obtaining cluster result includes：
The N number of class of initial construction, N is the quantity of wind power plant to be divided here；
The Euclidean distance of N number of sample between any two is calculated, distance matrix d is obtained；
Between class distance is calculated, it is determined that simultaneously combined distance most close 2 classes are 1 new class；
Whether the quantity for judging the class after cluster is 1, if the number of class is not equal to 1, between class distance is computed repeatedly, it is determined that and closing
And be 1 new class apart from 2 most close classes, until the quantity of the class after cluster is 1；
Obtain hierarchical clustering result.
5. according to the method described in claim 1, it is characterised in that described to be selected at least according to relative coefficient and predicated error
One wind power plant includes as the step of representing wind power plant：
In every sub-regions, the wind power plant of one relative coefficient of selection and prediction level all higher than average level is used as representative
Wind power plant.
6. a kind of prediction meanss of region wind-powered electricity generation short term power, it is characterised in that described device includes：
Region division module, is multiple by region division for being clustered to the wind power plant in region, and according to cluster result
Subregion；
Relative coefficient acquisition module, the correlation system exerted oneself with region wind power output for obtaining each wind power plant in region
Number；
Error acquisition module, the predicated error for obtaining each wind power plant in each subregion；
Wind power plant selecting module, in every sub-regions, at least one wind to be selected according to relative coefficient and predicated error
Electric field is as representing wind power plant；
Weight coefficient computing module, the weight coefficient of wind power plant is represented for being calculated according to the installed capacity of subregion；
Wind power prediction module, for representing wind farm power prediction value and weight coefficient, zoning wind-powered electricity generation according to each
Power prediction value.
7. device according to claim 6, it is characterised in that the region division module includes：
Spatial mode matrix acquiring unit, empirical orthogonal function decomposition is carried out for the historical power sequence to wind power plant in region,
Obtain spatial mode matrix；
Cluster cell, the spatial mode matrix for being obtained using hierarchical clustering method to empirical orthogonal function decomposition carries out cluster point
Analysis, obtains hierarchical clustering result；
Sub-zone dividing unit, for being some subregions by region division based on hierarchical clustering result.
8. device according to claim 7, it is characterised in that the spatial mode matrix acquiring unit is additionally operable to：
The historical power data sample of all wind power plants in region is obtained, raw data matrix X is obtained；
Raw data matrix X is standardized, normalized matrix Y is obtained；
Normalized matrix Y covariance matrix R, wherein R=YYT；
Calculating matrix R characteristic root, and, the characteristic vector of each characteristic root is obtained, as original by characteristic root in magnitude order
Each spatial mode of wind power plant, forms spatial mode matrix.
9. device according to claim 7, it is characterised in that cluster cell is additionally operable to：
The N number of class of initial construction, N is the quantity of wind power plant to be divided here；
The Euclidean distance of N number of sample between any two is calculated, distance matrix d is obtained；
Between class distance is calculated, it is determined that simultaneously combined distance most close 2 classes are 1 new class；
Whether the quantity for judging the class after cluster is 1, if the number of class is not equal to 1, between class distance is calculated again, it is determined that and closing
And be 1 new class apart from 2 most close classes, until the quantity of the class after cluster is 1；
Obtain hierarchical clustering result.
10. device according to claim 6, it is characterised in that the wind power plant selecting module is additionally operable to：
In every sub-regions, the wind power plant of one relative coefficient of selection and prediction level all higher than average level is used as representative
Wind power plant.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710295885.6A CN107194499A (en) | 2017-04-28 | 2017-04-28 | The Forecasting Methodology and prediction meanss of region wind-powered electricity generation short term power |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710295885.6A CN107194499A (en) | 2017-04-28 | 2017-04-28 | The Forecasting Methodology and prediction meanss of region wind-powered electricity generation short term power |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107194499A true CN107194499A (en) | 2017-09-22 |
Family
ID=59872391
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710295885.6A Pending CN107194499A (en) | 2017-04-28 | 2017-04-28 | The Forecasting Methodology and prediction meanss of region wind-powered electricity generation short term power |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107194499A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108335010A (en) * | 2017-12-18 | 2018-07-27 | 中国电力科学研究院有限公司 | A kind of wind power output time series modeling method and system |
| CN109063930A (en) * | 2018-08-30 | 2018-12-21 | 上海电力学院 | A kind of dynamic wind power plant general power prediction technique based on clustering |
| CN109800919A (en) * | 2019-01-30 | 2019-05-24 | 云南电网有限责任公司电力科学研究院 | A kind of regional wind power integrally active power forecasting method and device |
| CN110175639A (en) * | 2019-05-17 | 2019-08-27 | 华北电力大学 | A kind of short-term wind power forecast method based on Feature Selection |
| CN110288140A (en) * | 2019-06-14 | 2019-09-27 | 西北大学 | A kind of opioid drug spatial prediction technique based on geo-relevance model |
| CN110400019A (en) * | 2019-07-29 | 2019-11-01 | 南京工程学院 | A kind of wind farm power prediction optimization method based on data mining |
| CN110400011A (en) * | 2019-07-12 | 2019-11-01 | 国能日新科技股份有限公司 | Output of wind electric field declares the determination method and device of scheme in electric power spot exchange |
| CN110458341A (en) * | 2019-07-25 | 2019-11-15 | 山东大学 | A kind of ultra-short term wind-powered electricity generation prediction technique and system considering meteorological features |
| CN113792778A (en) * | 2021-09-07 | 2021-12-14 | 国网经济技术研究院有限公司 | Wind power plant partition and output correlation coefficient calculation method and system thereof |
| CN117111182A (en) * | 2023-10-19 | 2023-11-24 | 安徽省公共气象服务中心(安徽省突发公共事件预警信息发布中心) | Large fog short-term forecasting method based on air block reverse tracking |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120029712A1 (en) * | 2011-09-02 | 2012-02-02 | Onsemble, Inc. | Systems, methods and apparatus for indexing and predicting wind power output from virtual wind farms |
| CN102509028A (en) * | 2011-12-20 | 2012-06-20 | 国网电力科学研究院 | Whole network wind electric power forecasting algorithm for power grid based on technology of statistical rising scale |
| CN102570449A (en) * | 2011-12-21 | 2012-07-11 | 西安交通大学 | Regional wind power prediction method based on spatial and temporal distribution characteristics |
| CN102938562A (en) * | 2012-07-19 | 2013-02-20 | 中国电力科学研究院 | Prediction method of total wind electricity power in area |
| US20140074300A1 (en) * | 2012-09-07 | 2014-03-13 | Opower, Inc. | Thermostat Classification Method and System |
| CN106447234A (en) * | 2016-10-26 | 2017-02-22 | 国网电力科学研究院武汉南瑞有限责任公司 | A wind power plant abandoned wind power assessment method based on a hierarchical clustering method |
-
2017
- 2017-04-28 CN CN201710295885.6A patent/CN107194499A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120029712A1 (en) * | 2011-09-02 | 2012-02-02 | Onsemble, Inc. | Systems, methods and apparatus for indexing and predicting wind power output from virtual wind farms |
| CN102509028A (en) * | 2011-12-20 | 2012-06-20 | 国网电力科学研究院 | Whole network wind electric power forecasting algorithm for power grid based on technology of statistical rising scale |
| CN102570449A (en) * | 2011-12-21 | 2012-07-11 | 西安交通大学 | Regional wind power prediction method based on spatial and temporal distribution characteristics |
| CN102938562A (en) * | 2012-07-19 | 2013-02-20 | 中国电力科学研究院 | Prediction method of total wind electricity power in area |
| US20140074300A1 (en) * | 2012-09-07 | 2014-03-13 | Opower, Inc. | Thermostat Classification Method and System |
| CN106447234A (en) * | 2016-10-26 | 2017-02-22 | 国网电力科学研究院武汉南瑞有限责任公司 | A wind power plant abandoned wind power assessment method based on a hierarchical clustering method |
Non-Patent Citations (2)
| Title |
|---|
| 别朝红等: "《一种考虑时空分布特性的区域风电功率预测方法》", 《西安交通大学学报》 * |
| 陈颖等: "《基于统计升尺度方法的区域风电场群功率预测》", 《电力系统自动化》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108335010A (en) * | 2017-12-18 | 2018-07-27 | 中国电力科学研究院有限公司 | A kind of wind power output time series modeling method and system |
| CN109063930A (en) * | 2018-08-30 | 2018-12-21 | 上海电力学院 | A kind of dynamic wind power plant general power prediction technique based on clustering |
| CN109800919A (en) * | 2019-01-30 | 2019-05-24 | 云南电网有限责任公司电力科学研究院 | A kind of regional wind power integrally active power forecasting method and device |
| CN109800919B (en) * | 2019-01-30 | 2023-07-11 | 云南电网有限责任公司电力科学研究院 | Regional wind farm overall active power prediction method and device |
| CN110175639B (en) * | 2019-05-17 | 2021-06-11 | 华北电力大学 | Short-term wind power prediction method based on feature selection |
| CN110175639A (en) * | 2019-05-17 | 2019-08-27 | 华北电力大学 | A kind of short-term wind power forecast method based on Feature Selection |
| CN110288140A (en) * | 2019-06-14 | 2019-09-27 | 西北大学 | A kind of opioid drug spatial prediction technique based on geo-relevance model |
| CN110288140B (en) * | 2019-06-14 | 2023-04-07 | 西北大学 | Opioid spatial propagation prediction method based on geographical correlation model |
| CN110400011A (en) * | 2019-07-12 | 2019-11-01 | 国能日新科技股份有限公司 | Output of wind electric field declares the determination method and device of scheme in electric power spot exchange |
| CN110458341A (en) * | 2019-07-25 | 2019-11-15 | 山东大学 | A kind of ultra-short term wind-powered electricity generation prediction technique and system considering meteorological features |
| CN110400019A (en) * | 2019-07-29 | 2019-11-01 | 南京工程学院 | A kind of wind farm power prediction optimization method based on data mining |
| CN113792778A (en) * | 2021-09-07 | 2021-12-14 | 国网经济技术研究院有限公司 | Wind power plant partition and output correlation coefficient calculation method and system thereof |
| CN113792778B (en) * | 2021-09-07 | 2023-10-10 | 国网经济技术研究院有限公司 | Wind power plant partition and output correlation coefficient calculation method and system thereof |
| CN117111182A (en) * | 2023-10-19 | 2023-11-24 | 安徽省公共气象服务中心(安徽省突发公共事件预警信息发布中心) | Large fog short-term forecasting method based on air block reverse tracking |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107194499A (en) | The Forecasting Methodology and prediction meanss of region wind-powered electricity generation short term power | |
| Um et al. | Factor analysis and multiple regression between topography and precipitation on Jeju Island, Korea | |
| CN107909211B (en) | Wind field equivalent modeling and optimization control method based on fuzzy c-means clustering algorithm | |
| CN104331572A (en) | Wind power plant reliability modeling method considering correlation between air speed and fault of wind turbine generator | |
| CN106845642A (en) | A kind of self adaptation multi-target evolution method of belt restraining cloud workflow schedule | |
| CN106503279B (en) | A kind of modeling method for transient stability evaluation in power system | |
| CN107067099A (en) | Wind power probability forecasting method and device | |
| Rajaniemi et al. | Classifying gamma-ray bursts using self-organizing maps | |
| CN112132365B (en) | Multi-scale feature-based forecast wind speed correction method and system | |
| CN108667069A (en) | A kind of short-term wind power forecast method returned based on Partial Least Squares | |
| CN112017070A (en) | Method and system for evaluating transient stability of power system based on data enhancement | |
| CN103268573A (en) | Wind power plant marker post draught fan selection method based on principal component analysis | |
| CN111680823A (en) | Wind direction information prediction method and system | |
| CN107657349A (en) | A kind of reservoir power generation dispatching Rules extraction method by stages | |
| CN112328851B (en) | Distributed power supply monitoring method and device and electronic equipment | |
| CN110070227A (en) | Migration neural network power prediction method suitable for grid-connected photovoltaic power generation | |
| Mäkelä | Performance analysis of a process‐based stand growth model using Monte Carlo techniques | |
| CN116542030A (en) | Double-fed fan parameter identification method, system and equipment based on track sensitivity | |
| CN104200001B (en) | The choosing method of mark post blower fan | |
| CN110717623A (en) | Photovoltaic power generation power prediction method, device and equipment integrating multiple weather conditions | |
| Fetisova et al. | Determining the parameters of Weibull function to estimate the wind power potential in conditions of limited source meteorological data | |
| CN115833240A (en) | Capacity allocation optimization method of wind-light-water complementary system considering cascade hydropower station connection | |
| CN112990533B (en) | Wind power cluster power prediction method adopting sparse constraint and dynamic weight distribution | |
| Özkan et al. | A data mining-based wind power forecasting method: results for wind power plants in turkey | |
| CN110659843B (en) | Nuclear power conventional island water supply system reliability assessment method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170922 |