CN105320809A - Wind speed prediction method for wind farm spatial correlation - Google Patents

Wind speed prediction method for wind farm spatial correlation Download PDF

Info

Publication number
CN105320809A
CN105320809A CN201510640990.XA CN201510640990A CN105320809A CN 105320809 A CN105320809 A CN 105320809A CN 201510640990 A CN201510640990 A CN 201510640990A CN 105320809 A CN105320809 A CN 105320809A
Authority
CN
China
Prior art keywords
data
wind
wind speed
prediction
correlation coefficient
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201510640990.XA
Other languages
Chinese (zh)
Other versions
CN105320809B (en
Inventor
冯海林
赵玉宏
赵艳青
杨国平
齐小刚
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xidian University
Original Assignee
Xidian University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Xidian University filed Critical Xidian University
Priority to CN201510640990.XA priority Critical patent/CN105320809B/en
Publication of CN105320809A publication Critical patent/CN105320809A/en
Application granted granted Critical
Publication of CN105320809B publication Critical patent/CN105320809B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation

Landscapes

  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The present invention discloses a wind speed prediction method of a wind farm. Problems are mainly considered that spatial correlation among wind farms, unscented Kalman Filter optimization, and the like are not well considered in an existing method. The method mainly comprises: calculating a rank correlation coefficient between a target wind farm and the other 21 wind farms if 22 wind farms are given; determining a wind farm for prediction according to the correlation coefficient; and selecting a wind farm with a Kendall rank correlation coefficient greater than 0.55 and a Spearman rank correlation coefficient greater than 0.75; then establishing a non-linear state space model by using support vector machine regression, and performing unscented Kalman filter prediction by using the established non-linear state space model; optimizing a scale parameter of the unscented Kalman filter according to a principle of prediction error minimization; and finally, selecting wind speed data of a wind farm of a same time in four years, and performing grey correlation analysis by using the wind speed data and wind speed data of a target wind power turbine, No.9 wind power turbine, of the same time in the first year.

Description

A kind of wind speed forecasting method for wind energy turbine set spatial coherence
Technical field
Predicting wind speed of wind farm field of the present invention, particularly predicts at the comparatively intensive wind speed of area to a certain wind energy turbine set of wind energy turbine set distribution, ignores wind field position and the difficult problem of precision of prediction deficiency that causes for solving in forecasting wind speed process.
Background technology
Wind-power electricity generation, because have environmental protection, the plurality of advantages such as renewable, develops rapidly in recent years, has become the desirable energy that the whole world is generally acknowledged.But when wind-powered electricity generation penetrates after power reaches certain value, the randomness of wind-powered electricity generation, undulatory property and instability will have a huge impact the operation of electric system.Relevant scholar has carried out many correlative studys to this aspect, has drawn some conclusions be just of practical significance.And to reduce this impact just main will making wind speed or wind power exactly predict more accurately.
The topmost determinative of blower fan generated output is wind speed, so can as the key component of prediction wind power for effective prediction of wind speed.At present, the method both at home and abroad about forecasting wind speed generally can be divided into two classes: a class is the statistical method utilizing historical data modeling, and another kind of is utilize numerical weather forecast and topographic physical method.The former comprises, and time series is sent out, Kalman filtering method, neural network etc.Physical method generally comprises the forecast model, Spatial coherence method etc. that utilize numerical weather forecast.But wind has the feature of oneself as a kind of spontaneous phenomenon, existing method can not well be predicted wind speed.For the prediction of wind speed, not only to consider single wind series, need historical wind speed sequence and the real time data of wind energy turbine set, also must consider geographic position, roughness, wind direction, air pressure, the physical factor such as temperature and locus.
Therefore a good forecasting wind speed model, not only will consider the temporal correlation between wind speed, also will consider the spatial coherence of the wind speed of ambient wind electric field, could improve precision of prediction like this, improve forecast model.
In document Short-termwindpredictionusinganunscentedKalmanfilterbase dstate-spacesupportvectorregressionapproach, author gives a kind of wind speed forecasting method based on the electrodeless Kalman filtering of Nonlinear state space model, in order to set up the state equation of state-space model, the method of Support vector regression is have employed in literary composition, this method can carry out real-time prediction to wind speed, and the impact of wind speed randomness can be reduced, but, this method only considered single wind speed time series, wind field wind speed around it is not studied, it not a kind of complete Forecasting Methodology.
In document Applicationofartificialneuralnetworksforthewindspeedpred ictionofstationusingreferencestationsdata, author proposes and utilizes neural network model to predict wind speed, this process employs air speed data the choosing as input data of Wind Field, consider spatial coherence, but how the power for such as correlativity goes to differentiate in literary composition is not mentioned, and it not a kind of real-time Forecasting Methodology.
In document Ahybridstatisticalmethodtopredictwindspeedandwindpower, author proposes a kind of hybrid prediction model, in order to reduce fluctuations in wind speed to the impact predicted the outcome, author proposes to carry out wavelet decomposition to raw data, sub-sequences carries out time series forecasting, finally merges predicting the outcome of each subsequence again.The method does not still consider its spatial coherence.
In document Windpowerpredictionbasedonnumericalandstatisticalmodel, the method is predicted wind speed according to numerical weather forecast and Kalman prediction model, author considers the impact of meteorologic parameter on wind speed, and utilize Kalman filter model to carry out real-time prediction to wind speed, but due to this state-space model be linear, so the actual conditions of wind speed well can not be portrayed.
In forecasting wind speed process, Time and place correlativity between wind speed is and deposits, and it is different along with the difference of distance, except document Applicationofartificialneuralnetworksforthewindspeedpred ictionofstationusingreferencestationsdata considers the correlativity of wind speed between Wind Field in above-mentioned document, remaining does not all consider the spatial coherence between wind speed, larger impact is certainly existed on precision of prediction, so forecast model can be perfect further.
Summary of the invention
The object of the invention is to the deficiency not considering its spatial coherence for current most of forecasting wind speed model, a kind of wind speed hybrid prediction model is proposed, first this model compares the power of spatial coherence between its each wind field based on copula function, then utilize support vector machine to set up Nonlinear state space model, finally utilize the Unscented kalman filtering of optimization to predict wind speed.The real-time estimate of this model realization to wind speed, and consider temporal correlation and the spatial coherence of wind speed.Before introducing the present invention, first provide 2 definition.
The index utilizing copula function to calculate spatial coherence between two wind fields general is kendall rank correlation coefficient and Spearman rank correlation coefficient.We provide definition and the computing method of these two rank correlation coefficients below.
Definition 1 (Kendall rank correlation coefficient) makes (x 1, y 1) and (x 2, y 2) be independent identically distributed stochastic variable, definition
τ≡P[(x 1-x 2)(y 1-y 2)＞0]-P[(x 1-x 2)(y 1-y 2)＜0]
＝2P[(x 1-x 2)(y 1-y 2)＞0]-1
For kendall rank correlation coefficient, be designated as τ.
If stochastic variable X, the marginal distribution of Y is respectively F (x), G (y), and corresponding copula function is C (u, v), wherein u=F (x), v=G (y), u, v ∈ [0,1], then kendall rank correlation coefficient τ can have corresponding copula function C (u, v) to provide
Definition 2 (Spearman rank correlation coefficients) make (x 1, y 1), (x 2, y 2) and (x 3, y 3) be independent identically distributed stochastic variable, definition
ρ≡3{P[(x 1-x 2)(y 1-y 3)＞0]-P[(x 1-x 2)(y 1-y 3)＜0]}
For Spearman rank correlation coefficient, be designated as ρ.
If stochastic variable X, the marginal distribution of Y is respectively F (x), G (y), and corresponding copula function is C (u, v), wherein u=F (x), v=G (y), u, v ∈ [0,1], then kendall rank correlation coefficient τ can have corresponding copula function C (u, v) to provide
Note: when calculating this two rank correlation coefficients, the copula function that the present invention adopts is normal distyribution function and T distribution function.Data set used in the present invention is from the air speed data (on Dec 31,1 day to 2009 January in 2006, every 1 hour record once) of public data collection Wisconsin, USA (Wisconsinstate, WI) 22 wind energy turbine set.
Select step during input data as follows using rank correlation coefficient:
1: from data centralization select on Dec 30,24 days to 2009 Dec in 2009 data of totally 7 days.
2: the validity detecting data, fill missing data, the present invention utilizes the mean value of around it 1 day to replace missing data.
3: test of normality is carried out to the data detected, then ask its empirical distribution function, kernel distribution estimator is carried out to its raw data, just can carry out parameter estimation to copula function afterwards.
4: the kendall rank correlation coefficient and the Spearman rank correlation coefficient that calculate its correspondence according to Normcopula and the tcopula function calculated.
5: the size according to the kendall rank correlation coefficient calculated and Spearman rank correlation coefficient chooses suitable input data.
The rank correlation coefficient table that we calculate according to said method is in table 1.
Kendall_norm Spearman_norm Kendallt Spearman_t Kendall Spearman
0.5031 0.6937 0.5530 0.7480 0.5901 0.7566
0.5548 0.7499 0.5970 0.7924 0.6090 0.7592
0.4550 0.6377 0.4890 0.6776 0.5021 0.6463
0.5110 0.7026 0.5497 0.7445 0.5813 0.7365
0.5333 0.7271 0.5738 0.7695 0.5713 0.7307
0.5528 0.7478 0.5874 0.7831 0.6226 0.7778
0 0.5532 0.7483 0.5915 0.7870 0.6162 0.7742
1 0.4354 0.6139 0.4715 0.6572 0.4978 0.6693
4 0.5304 0.7239 0.5700 0.7656 0.5798 0.7382
6 0.3280 0.4754 0.3731 0.5352 0.3750 0.4666
7 0.5528 0.7478 0.6042 0.7994 0.6370 0.7837
8 0.1956 0.3564 0.3215 0.4235 0.1674 0.2014
9 0.5767 0.7724 0.6185 0.8128 0.6456 0.7900
0 0.0987 0.1477 0.2347 0.3460 0.0966 0.0972
1 0.5238 0.6937 0.5984 0.7689 0.6015 0.7215
2 0.5598 0.7551 0.5998 0.7952 0.6341 0.8010
Table 1
By selection and comparison, we filter out the blower fan that No. 3, No. 4, No. 5, No. 6, No. 7, No. 11, No. 14 and No. 19 blower fans input as data.
Support vector machine is a kind of machine learning techniques, and support vector regression is the prediction regression model utilizing Non-linear Kernel function and support vector to set up specially.In support vector regression, most crucial step finds a function f ∈ F (F is a collection of functions), corresponding expected risk function is allowed to reach minimum value, that is: R [f]=∫ l (y-f (x)) dP (x, y).Wherein l () represent loss function, represent the deviation between y and f (x), common type be l ()=| y-f (x) | p, wherein p is certain positive integer.
The basic thought of SVR is as follows:
For given training sample { (x 1, y 1), (x 2, y 2) ..., (x n, y n) wherein if regression function is
Above-mentioned regression problem can be in the hope of according to Lagrange's theorem and KKT condition:
Wherein for kernel function.X r, x sfor identification vector.
To data during 24 days 08 Dec in 2009 with train for the air speed data of No. 9 blower fans, the α obtained during the 2009 year Dec 24 day 00 of this method according to 8 blower fans selecting 1:8, in table 2.
α α *
0.3536 0
0.3424 0.3424
0 0.3536
0 0.3536
0 0.3536
0.2779 0.2779
0.3536 0
0.3536 0
Table 2
This method is to the optimizing process following steps of Unscented kalman filtering:
1: the feasible set of specifying a scale parameter κ, minimum value is 0, and maximal value is selected according to document.
2: select the κ that initial 0, this method is got
3: order bring in the step of Unscented kalman filtering to calculate and predict the outcome.
4: if
Then get κ j+1=κ j+, otherwise κ j+1=κ j.
5: circulation 3-4 walks, until the threshold value that predicated error reaches setting stops.
The present invention finally utilizes grey incidence coefficient further to optimize input data again, and optimization method is as follows:
1: for 8 wind fields chosen, select 2006-2009 annual Dec 24 to data on Dec 30 respectively, each wind field selects 4 groups of data, often organizes 168 data.
2: the group data of each wind field and No. 9 wind fields data on Dec 30,24 days to 2009 Dec in 2009 are carried out grey relational grade calculating.
3: according to the size of grey incidence coefficient, select corresponding input data.
What the present invention selected is Absolute Correlation Analysis, and the account form of this degree of association is as follows:
1: by original series X 0={ x 0(k), k=1,2 ..., n} and X i={ x i(k), k=1,2 ..., n} carries out just value process:
2 calculate x 0with x iabsolute Correlation Analysis be:
Wherein,
Result of calculation is in table 3.
0.6736 0.7399 0.7726 0.7307 0.7348 0.6674 0.7134 0.6789
0.7371 0.7769 0.7848 0.7788 0.7946 0.7660 0.7765 0.7068
0.7151 0.8041 0.7995 0.7697 0.8238 0.7582 0.7598 0.7411
0.7805 0.8056 0.7801 0.7783 0.7749 0.7752 0.8033 0.7201
Table 3
Relative to prior art, the present invention possesses following advantage:
(1) the present invention effectively can improve precision of prediction and to compare general traditional prediction method.
(2) the present invention not only achieves the real-time estimate of wind speed, and considers the spatial coherence between wind field.
(3) the present invention not only considers the spatial coherence of same year wind field data, and considers the temporal correlation between annual each wind speed.For shortage of data process and utilize the degree of association to judge the Similarity measures of different year with time segment data, compare the similarity between same time segment data, improve precision of prediction.
(4) the present invention not only considers the correlativity between data in forecasting process, and what set up is nonlinear model, reduces the problem of the precision of prediction deficiency caused due to wind speed randomness, undulatory property preferably.
(5) taken into full account the spatial coherence between wind field, more tallied with the actual situation.
(6) go to predict wind speed by a nonlinear Forecasting Methodology, utilize Unscented kalman filtering to go to predict wind speed, can ensure its real-time predicted, and more meet the actual rule of wind speed, result is more accurate.
(7) calculating of grey Absolute data relating extent is carried out to 4 years air speed datas of selected wind field and target wind field air speed data, wind speed mutation can be reduced on the impact predicted the outcome, namely adopt the data stronger with prediction period similarity to go to predict, precision of prediction is higher.
(8) carry out pre-service to data can avoid in subsequent steps due to predicted impact that data aspect causes.
Accompanying drawing explanation
Fig. 1 is general flow chart of the present invention;
Fig. 2 is the scale parameter Optimization Steps figure of Unscented kalman filtering;
Fig. 3 utilizes grey incidence coefficient to select optimum input traffic journey figure;
Fig. 4 is the location map of each wind energy turbine set;
Fig. 5 is the wind speed curve variation diagram of 3 wind energy turbine set;
Fig. 6 is Ken Deer rank correlation coefficient and the selected wind field fitted figure to target wind field distance;
Fig. 7 is the fitted figure of Si Pier rank correlation coefficient;
Fig. 8 does not select through grey relational grade, carries out the wind speed result figure predicted by same year data;
Fig. 9 is the histogram of Fig. 8 prediction residual;
Figure 10 is the figure that predicts the outcome of hybrid algorithm;
Figure 11 utilizes AR-KALMAN filter forecasting result figure;
Figure 12 is that wavelet neural network method predicts the outcome figure;
Figure 13 is that hybrid algorithm predicts the outcome residual plot;
Figure 14 is AR-KALMAN filter forecasting result residual plot;
Figure 15 is that wavelet neural network method predicts the outcome residual plot;
Embodiment
For enabling the expression that target of the present invention, technical scheme and advantage are clearly clear and definite, in conjunction with above-mentioned accompanying drawing, detailed implementation step of the present invention is described further.
With reference to figure 1, concrete steps of the present invention are as follows:
Step 1. selects raw data, concentrates choose each wind field data from public data, the annual beginning on Dec 24 of data acquisition 2006 ~ 2009 these 4 years, Dec 30 end of day.
Step 2. carries out disappearance inspection to all data collected, and the wind field many to missing data is directly deleted, and the remaining averaging method that utilizes is filled, that is: suppose a ifor missing data, then the data being filled into this position are
If a ibefore and after 12 data also have recursion of drawing back forward during disappearance, until obtain 12, front and back data; If a ibe first data, now in like manner, a iduring for last data, if a ifor before data centralization 12 and rear 12 data time, in order to convenience of calculation, we still select with head and the tail fill method consistent.Result can show, is not too large on the impact of predicated error.
Step 3. obtains except No. 1 according to the padding scheme of top, No. 8, No. 12, No. 13, the complete data set of No. 15 wind field other wind fields outer.
Step 4. selects input data, specific as follows: the rank correlation coefficient calculating No. 9, target wind field and other 16 wind fields the 4th annual datas according to copula function, and the size according to rank correlation coefficient selects input wind field.Through relatively more selected 8 groups of wind fields.
Step 5. selects selected wind field 2006 ~ 2009 data, and data acquisition is from annual Dec 24 to Dec 30, and totally 32 groups of data, often organize 168 data.Target wind field image data is on Dec 30,24 days to 2009 Dec in 2009, totally 168 data.
Step 6. calculates the grey relational grade of the annual data of each wind field and target wind field the 4th annual data.
Step 7. selects final input data according to the magnitude relationship of grey relational grade.
Step 8. carries out modeling analysis to the input data chosen, and sets up Nonlinear state space model according to SVR.
Step 9. carries out Unscented kalman filtering prediction according to the state-space model established.
Step 10. is optimized the scale parameter in Unscented kalman filtering according to error minimum principle
Step 11. obtains optimum predicting the outcome.
Effect of the present invention can be further described by following emulation:
1. simulated conditions
The present invention, by carrying out the experiment simulation of distinct methods to same data set, illustrates the validity of algorithm.Simulation laboratory is at a 4G internal memory, and Celeron double-core 2.6Hz, 32 win7 operating systems, use MATLAB2012b to carry out.
2. emulate content
Emulation 1, the data set choosing 22 wind energy turbine set of public data collection Wisconsin, USA is tested.This dataset acquisition time is every 1 hour once, on Dec 31, in time from 1 day to 2009 January in 2006.Data have 4 attributes, are respectively T, dir, spd and airmp.Based on method herein, the attribute that the present invention chooses when emulating only has wind speed.Data set D1 is longitude and the dimension of 22 wind energy turbine set.
Fig. 4 is the 3 dimension location maps drawn according to the longitude of each wind energy turbine set and dimension.This figure is with the centre of sphere of the earth for spherical coordinates initial point, and earth radius is the radius of a ball, according to the location map drawn through dimension, can find out the range distribution of each wind energy turbine set and target wind field in figure clearly.Fig. 5 is the change curve of 3 same time period wind speed of wind energy turbine set, and its culminant star line is a wind field, and dotted line is No. 2 wind fields, and dotted line is No. 3 wind fields, and data amount check is 92.Can find out that the wind speed of each wind field same time period has very strong similarity.
Emulation 2, data set D2 is populated complete data set, comprises the data in 17 wind fields 2006 ~ 2009 annual Dec 24 to Dec 30.Data set D3 is the air speed data in each wind energy turbine set on Dec 30th, 24 days 1 Dec in 2009 of extracting from D2.Fig. 6 is Ken Deer rank correlation coefficient and the selected wind field fitted figure to target wind field distance, and straight line is represented as matched curve, and loose point is actual value.Fig. 7 is the fitted figure of Si Pier rank correlation coefficient.
Emulation 3, adopts data set D3 to detect algorithm of the present invention, and Fig. 8 does not select through grey relational grade, and directly carry out the wind speed result figure predicted by same year data, Fig. 9 is the histogram of prediction residual.Prediction step number is 24 steps, i.e. the wind speed in target wind field on Dec 31st, 2009, and dotted line is true wind speed, and star line is predicted value.
Emulation 4, in order to improve precision of prediction better, we need to carry out similarity selection to data, namely utilize grey relational grade using select each wind field in 4 years data and the 4th year wind field same time period of target wind field relevance the strongest that year data as inputting data, through the method select data set we be called data set D3 '.When Figure 10 is D3 ' conduct input data, the figure that predicts the outcome of this hybrid algorithm, Figure 11 and Figure 12 is respectively the result figure utilizing AR-KALMAN filtering and wavelet neural network method to predict same data, and Figure 13, Figure 14 and Figure 15 are the histograms of the prediction residual of above-mentioned three kinds of methods.
Symbol description
SVR: Support vector regression
T: temperature
Dir: wind direction
Spd: wind speed
Airmp: air pressure
D1: emulated data collection 1
D2: emulated data collection 2
D3: emulated data collection 3.

Claims (4)

1. utilize spatial coherence to a method for predicting wind speed of wind farm, it is characterized in that: it comprises following step
Rapid:
S1: select raw data, in disclosed wind farm data collection image data, carries out shortage of data inspection to the data collected and supplements complete;
S2: according to copula function, its rank correlation coefficient is calculated to supplementary complete data, selects the wind field of suitable input data according to the size of rank correlation coefficient;
S3: carry out the calculating of grey Absolute data relating extent to 4 years air speed datas of selected wind field and target wind field air speed data, according to the final input data of the magnitude relationship Confirming model of grey Absolute data relating extent;
S4: carry out modeling analysis to the final input data chosen, sets up state equation and the measurement equation of Nonlinear state space model according to support vector regression, and carries out estimation prediction by Unscented kalman filtering to the state of model;
S5: according to the choice criteria of setting, the scale parameter in Unscented kalman filtering be optimized and upgrade, being predicted the outcome.
2. a kind of spatial coherence that utilizes according to claim 1 is to the method for predicting wind speed of wind farm, it is characterized in that: the detailed process of described step S1 is: the data of the same time period selecting each wind energy turbine set annual; 22 groups that select altogether, often organize 192 data, then carry out disappearance inspection to all data collected and directly delete for the wind field of missing data more than 30, the remaining averaging method that utilizes is filled, that is, suppose a ifor missing data, be then filled into the data of this position
If a ibefore and after 12 data when also having a disappearance, recursion of drawing back forward, until obtain 12, front and back data; If a ibe first data, then now in like manner, a iduring for last data, if a ifor before data centralization 12 and rear 12 data time, still select and from beginning to end identical fill method; And then obtain except No. 1, and No. 8, No. 12, No. 13, the complete data set of No. 15 wind field other wind fields outer.
3. a kind of spatial coherence that utilizes according to claim 2 is to the method for predicting wind speed of wind farm, it is characterized in that: described step S5 to be optimized the scale parameter in Unscented kalman filtering according to error minimum principle to be predicted the outcome.
4. a kind of spatial coherence that utilizes according to claim 3 is to the method for predicting wind speed of wind farm, it is characterized in that: the detailed process of described step S5 is as follows:
Specify the feasible set λ ∈ [0,12] of a scale parameter λ, update method is as follows:
(1) initial λ is selected;
Getting it is wherein λ max=12, λ min=0
(2) when upgrading, at every turn at original λ jbasis on add a random value e j, this value meets normal state change, and expect to be 0, variance is very little;
κ j+=κ j+ e j make j=0,1,
(3) by above-mentioned λ j+and λ jsubstitute in Unscented kalman filtering respectively and carry out prediction and calculation;
(4) predicated error both calculating, gets the little person of error and enters next step renewal;
If
Then get λ j+1=λ j+
Otherwise λ j+1=λ j
(5) circulation 2-4 walks, until predicated error reaches the threshold value of setting or update times when reaching established standards, upgrades and stops, obtaining optimum scale parameter λ.
CN201510640990.XA 2015-09-30 2015-09-30 A kind of wind speed forecasting method for wind power plant spatial coherence Active CN105320809B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510640990.XA CN105320809B (en) 2015-09-30 2015-09-30 A kind of wind speed forecasting method for wind power plant spatial coherence

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510640990.XA CN105320809B (en) 2015-09-30 2015-09-30 A kind of wind speed forecasting method for wind power plant spatial coherence

Publications (2)

Publication Number Publication Date
CN105320809A true CN105320809A (en) 2016-02-10
CN105320809B CN105320809B (en) 2019-01-25

Family

ID=55248189

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510640990.XA Active CN105320809B (en) 2015-09-30 2015-09-30 A kind of wind speed forecasting method for wind power plant spatial coherence

Country Status (1)

Country Link
CN (1) CN105320809B (en)

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105824987A (en) * 2016-03-09 2016-08-03 浙江大学 Wind field characteristic statistical distributing model building method based on genetic algorithm
CN108090614A (en) * 2017-12-18 2018-05-29 哈尔滨工业大学 A kind of space wind field prediction model method for building up based on related coefficient
CN109032845A (en) * 2018-05-04 2018-12-18 马晨光 A kind of circuit and method of intelligent control power-off
CN109102101A (en) * 2017-06-21 2018-12-28 北京金风科创风电设备有限公司 Wind speed prediction method and system for wind power plant
CN109166110A (en) * 2018-08-15 2019-01-08 安徽省农业科学院土壤肥料研究所 A kind of intelligence vegetables nursery breeding automatic load system and packing method
CN109726802A (en) * 2018-12-29 2019-05-07 中南大学 A kind of railway and wind power plant environment wind speed machine learning prediction technique
CN109902863A (en) * 2019-02-15 2019-06-18 浙江财经大学 A kind of wind speed forecasting method and device based on multifactor temporal correlation
CN110704406A (en) * 2019-08-30 2020-01-17 珠海格力电器股份有限公司 Energy data processing method, device and equipment
CN111027852A (en) * 2019-12-09 2020-04-17 华南理工大学广州学院 Method for analyzing wind power generation benefits based on big data processing and grey theory
CN112052466A (en) * 2020-08-28 2020-12-08 西安电子科技大学 Support vector machine user data prediction method based on multi-party secure computing protocol
CN113591957A (en) * 2021-07-21 2021-11-02 国网上海市电力公司 Wind power output short-term rolling prediction and correction method based on LSTM and Markov chain

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101871948B (en) * 2010-06-09 2012-10-17 中国科学院深圳先进技术研究院 System and method for predicting wind speed of wind farm
KR101214841B1 (en) * 2011-11-22 2012-12-24 서강대학교산학협력단 Real time optimization system for a fixed-bed adsorption process and method therefor
CN102768738A (en) * 2012-07-31 2012-11-07 华锐风电科技(集团)股份有限公司 Method and device for power predication of wind power station
CN103605908A (en) * 2013-12-03 2014-02-26 天津工业大学 Wind speed sequence forecasting method based on Kalman filtering

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105824987A (en) * 2016-03-09 2016-08-03 浙江大学 Wind field characteristic statistical distributing model building method based on genetic algorithm
CN109102101B (en) * 2017-06-21 2021-01-26 北京金风科创风电设备有限公司 Wind speed prediction method and system for wind power plant
CN109102101A (en) * 2017-06-21 2018-12-28 北京金风科创风电设备有限公司 Wind speed prediction method and system for wind power plant
CN108090614A (en) * 2017-12-18 2018-05-29 哈尔滨工业大学 A kind of space wind field prediction model method for building up based on related coefficient
CN108090614B (en) * 2017-12-18 2021-05-18 哈尔滨工业大学 Method for establishing space wind field prediction model based on correlation coefficient
CN109032845A (en) * 2018-05-04 2018-12-18 马晨光 A kind of circuit and method of intelligent control power-off
CN109166110A (en) * 2018-08-15 2019-01-08 安徽省农业科学院土壤肥料研究所 A kind of intelligence vegetables nursery breeding automatic load system and packing method
CN109726802A (en) * 2018-12-29 2019-05-07 中南大学 A kind of railway and wind power plant environment wind speed machine learning prediction technique
CN109902863A (en) * 2019-02-15 2019-06-18 浙江财经大学 A kind of wind speed forecasting method and device based on multifactor temporal correlation
CN110704406A (en) * 2019-08-30 2020-01-17 珠海格力电器股份有限公司 Energy data processing method, device and equipment
CN111027852A (en) * 2019-12-09 2020-04-17 华南理工大学广州学院 Method for analyzing wind power generation benefits based on big data processing and grey theory
CN111027852B (en) * 2019-12-09 2023-04-18 华南理工大学广州学院 Method for analyzing wind power generation benefits based on big data processing and grey theory
CN112052466A (en) * 2020-08-28 2020-12-08 西安电子科技大学 Support vector machine user data prediction method based on multi-party secure computing protocol
CN113591957A (en) * 2021-07-21 2021-11-02 国网上海市电力公司 Wind power output short-term rolling prediction and correction method based on LSTM and Markov chain
CN113591957B (en) * 2021-07-21 2023-10-27 国网上海市电力公司 Wind power output short-term rolling prediction and correction method based on LSTM and Markov chain

Also Published As

Publication number Publication date
CN105320809B (en) 2019-01-25

Similar Documents

Publication Publication Date Title
CN105320809A (en) Wind speed prediction method for wind farm spatial correlation
Wang et al. Comparison of seven methods for determining the optimal statistical distribution parameters: A case study of wind energy assessment in the large-scale wind farms of China
CN110598290B (en) Method and system for predicting future hydropower generation capacity of basin considering climate change
Kwon Uncertainty analysis of wind energy potential assessment
CN101661530B (en) Method for acquiring steady-state equivalent wind speed and generated power in wind power station based on correlation analysis
CN103117546B (en) A kind of Ultrashort-term slide prediction method for wind power
CN114092832B (en) High-resolution remote sensing image classification method based on parallel hybrid convolutional network
Chen et al. A new method for identification of flood seasons using directional statistics
CN102495919B (en) Extraction method for influence factors of carbon exchange of ecosystem and system
CN110942194A (en) Wind power prediction error interval evaluation method based on TCN
CN104036121B (en) Wind measurement data wind speed correction method based on probability distribution transfer
Long et al. Reconstruction of historical arable land use patterns using constrained cellular automata: A case study of Jiangsu, China
CN104951834A (en) LSSVM (least squares support vector machine) wind speed forecasting method based on integration of GA (genetic algorithm) and PSO (particle swarm optimization)
CN105574615A (en) Spatial correlation and genetic algorithm (GA) based wind power forecast method for wavelet-BP neural network
CN107704992A (en) The method and device of transmission line lightning stroke risk assessment
CN103279813A (en) Steam load prediction method
CN106127330A (en) Fluctuating wind speed Forecasting Methodology based on least square method supporting vector machine
CN105303250A (en) Wind power combination prediction method based on optimal weight coefficient
CN109299208A (en) Transmission tower intelligent visual methods of risk assessment under a kind of typhoon disaster
CN112100911A (en) Solar radiation prediction method based on deep BISLTM
Siqueira et al. Generation of daily solar irradiation by means of artificial neural net works
CN104835073A (en) Unmanned aerial vehicle control system operation performance evaluating method based on intuitionistic fuzzy entropy weight
CN106886564A (en) A kind of method and device that NWP wind energy collection of illustrative plates is corrected based on space clustering
Tai et al. The development of a terrain-resolving scheme for the forward model and its adjoint in the four-dimensional Variational Doppler Radar Analysis System (VDRAS)
CN103268416B (en) Wind speed prediction method based on STCP-BP

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant