CN107292439A - A kind of method and apparatus for the short-term wind speed forecasting that Copula functions are mixed based on time-varying - Google Patents
A kind of method and apparatus for the short-term wind speed forecasting that Copula functions are mixed based on time-varying Download PDFInfo
- Publication number
- CN107292439A CN107292439A CN201710485792.XA CN201710485792A CN107292439A CN 107292439 A CN107292439 A CN 107292439A CN 201710485792 A CN201710485792 A CN 201710485792A CN 107292439 A CN107292439 A CN 107292439A
- Authority
- CN
- China
- Prior art keywords
- value
- wind speed
- probability density
- copula functions
- time
- 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
- 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 embodiment of the invention discloses a kind of method and apparatus for the short-term wind speed forecasting that Copula functions are mixed based on time-varying, according to the actual value and corresponding predicted value of historical wind speed data, time-varying mixing Copula functions are built；Using expectation maximization maximum likelihood method, determine that time-varying mixes the targeted parameter value of Copula functions.In the case of given wind speed value, Copula functions are mixed using time-varying, the conditional probability density function of forecasting wind speed error can be calculated；Conditional probability density function is converted into discrete conditional probability density function, and area accumulation is integrated to discrete conditional probability density function, the forecasting wind speed confidential interval under default confidence level may finally be obtained.It can be seen that, by above-mentioned technical proposal, the interval i.e. forecasting wind speed confidential interval of corresponding wind speed probabilistic forecasting can be obtained according to given confidence level, the probabilistic forecasting to wind speed is realized, effectively embodies the unascertained information of wind speed.
Description
Technical field
It is more particularly to a kind of that the short-term of Copula functions is mixed based on time-varying the present invention relates to technical field of wind power generation
The method and apparatus of forecasting wind speed.
Background technology
The randomness of wind speed and it is intermittent bring huge challenge to large-scale wind power is grid-connected, so wind speed is short-term pre-
The operation surveyed to power system is significant.
In the last few years, prediction of many methods to solve wind speed and wind power was proposed to this scholar studied to ask
Topic, but most research is all the deterministic forecast about wind speed and wind power, and to wind speed and the probability of wind power
Property forecasting research is relatively fewer.And compared to deterministic forecast, wind speed uncertainty can more be embodied to the probabilistic forecasting of wind speed
Information, is more beneficial for the operation and planning of power network.
It can be seen that, the probabilistic forecasting to wind speed how is realized, is those skilled in the art's urgent problem to be solved.
The content of the invention
The purpose of the embodiment of the present invention is to provide a kind of side for the short-term wind speed forecasting that Copula functions are mixed based on time-varying
Method and device, it is possible to achieve to the probabilistic forecasting of wind speed, effectively embody the unascertained information of wind speed.
In order to solve the above technical problems, the embodiment of the present invention provides a kind of short-term wind that Copula functions are mixed based on time-varying
The method of speed prediction, including：
According to the actual value and corresponding predicted value of historical wind speed data, time-varying mixing Copula functions are built；
Using expectation maximization-maximum likelihood method, determine that the time-varying mixes the targeted parameter value of Copula functions；
The targeted parameter value includes the parameter value and corresponding weight coefficient of every kind of Copula functions；
According to the predicted value and time-varying mixing Copula functions, the conditional probability for calculating forecasting wind speed error is close
Spend function；
The conditional probability density function is converted into discrete conditional probability density function, and to the discrete condition
Probability density function is integrated area accumulation, obtains the forecasting wind speed confidential interval under default confidence level.
Optionally, the actual value and corresponding predicted value according to air speed data, builds time-varying mixing Copula functions
Including：
Obtain the actual value and corresponding predicted value of historical wind speed data；
Calculate the actual value and the respective marginal distribution function of the predicted value；
Using experience distribution, the marginal distribution function is converted into uniformly distributed function；
According to the uniformly distributed function, time-varying mixing Copula functions are built.
Optionally, the utilization expectation maximization-maximum likelihood method, determines the time-varying mixing Copula functions
Targeted parameter value includes：
Using maximum likelihood estimate, the parameter value of every kind of Copula functions is determined；
Using expectation maximization method, the weight coefficient of every kind of Copula functions is determined.
Optionally, it is described according to the predicted value and time-varying mixing Copula functions, calculate forecasting wind speed error
Conditional probability density function include：
Copula functions are mixed using the time-varying, the joint distribution function of the actual value and the predicted value is set up；
Local derviation is asked to the joint distribution function, corresponding joint probability density function is obtained；
According to the second predicted value and the joint probability density function, the condition C opula density of the second actual value is calculated
Function；
According to predicated error, the condition C opula density functions of second actual value are converted into the predicated error
Conditional probability density function；The predicated error is determined according to the actual value and the predicted value.
Optionally, it is described that area accumulation is integrated to the discrete conditional probability density function, obtain pre-seting
Forecasting wind speed confidential interval under letter level includes：
Obtain the corresponding probable value of various discrete value of the discrete conditional probability density function；
The probable value is added up successively, corresponding accumulated value is got；
According to the accumulated value and the default confidence level, corresponding higher limit and lower limit are obtained；
According to the higher limit and the lower limit, forecasting wind speed confidential interval is determined.
The embodiment of the present invention additionally provides a kind of device for the short-term wind speed forecasting that Copula functions are mixed based on time-varying, bag
Include construction unit, determining unit, computing unit and obtain unit,
The construction unit, for the actual value and corresponding predicted value according to historical wind speed data, builds time-varying mixing
Copula functions；
The determining unit, for utilizing expectation maximization-maximum likelihood method, determines the time-varying mixing Copula
The targeted parameter value of function；The targeted parameter value includes the parameter value and corresponding weight coefficient of every kind of Copula functions；
The computing unit, for according to the predicted value and time-varying mixing Copula functions, calculating wind speed pre-
Survey the conditional probability density function of error；
It is described to obtain unit, for the conditional probability density function to be converted into discrete conditional probability density function,
And area accumulation is integrated to the discrete conditional probability density function, obtain the forecasting wind speed under default confidence level
Confidential interval.
Optionally, the construction unit includes obtaining subelement, computation subunit, transforming subunit and sets up subelement,
The acquisition subelement, actual value and corresponding predicted value for obtaining historical wind speed data；
The computation subunit, for calculating the actual value and the respective marginal distribution function of the predicted value；
The transforming subunit, for utilizing experience distribution, the marginal distribution function is converted into and is uniformly distributed letter
Number；
It is described to set up subelement, for according to the uniformly distributed function, building time-varying mixing Copula functions.
Optionally, the determining unit determines every kind of Copula functions specifically for utilizing maximum likelihood estimate
Parameter value；Using expectation maximization method, the weight coefficient of every kind of Copula functions is determined.
Optionally, the computing unit includes setting up subelement, obtaining subelement, computation subunit and transforming subunit,
It is described to set up subelement, for mixing Copula functions using the time-varying, set up the actual value and described pre-
The joint distribution function of measured value；
It is described to obtain subelement, for seeking local derviation to the joint distribution function, obtain corresponding joint probability density letter
Number；
The computation subunit, for according to the second predicted value and the joint probability density function, calculating second real
The condition C opula density functions of actual value；
The transforming subunit, for according to predicated error, by the condition C opula density functions of second actual value
It is converted into the conditional probability density function of the predicated error；The predicated error is true according to the actual value and the predicted value
It is fixed.
Optionally, the unit that obtains includes obtaining subelement, obtains subelement and determination subelement,
The acquisition subelement, the various discrete value for obtaining the discrete conditional probability density function is corresponding general
Rate value；
The acquisition subelement is additionally operable to add up to the probable value successively, gets corresponding accumulated value；
It is described to obtain subelement, for according to the accumulated value and the default confidence level, obtaining corresponding higher limit
And lower limit；
The determination subelement, for according to the higher limit and the lower limit, determining forecasting wind speed confidential interval.
Actual value and corresponding predicted value it can be seen from above-mentioned technical proposal according to historical wind speed data, during structure
Become mixing Copula functions；The parameter value of every kind of Copula functions that is included in time-varying mixing Copula functions and corresponding
Weight coefficient belongs to unknown target component, it is possible to use expectation maximization-maximum likelihood method, determines the time-varying mixing
The targeted parameter value of Copula functions.In the case of given wind speed value, Copula functions are mixed using the time-varying, can
To calculate the conditional probability density function of forecasting wind speed error；The conditional probability density function is converted into discrete condition
Probability density function, and area accumulation is integrated to the discrete conditional probability density function, it may finally obtain pre-
If the forecasting wind speed confidential interval under confidence level.It can be seen that, by above-mentioned technical proposal, it can be obtained according to given confidence level
Interval to corresponding wind speed probabilistic forecasting is forecasting wind speed confidential interval, realizes the probabilistic forecasting to wind speed, effectively
The unascertained information for embodying wind speed.
Brief description of the drawings
In order to illustrate the embodiments of the present invention more clearly, the required accompanying drawing used in embodiment will be done simply below
Introduce, it should be apparent that, drawings in the following description are only some embodiments of the present invention, for ordinary skill people
For member, on the premise of not paying creative work, other accompanying drawings can also be obtained according to these accompanying drawings.
Fig. 1 is a kind of method for the short-term wind speed forecasting that Copula functions are mixed based on time-varying provided in an embodiment of the present invention
Flow chart；
Fig. 2 is a kind of device for the short-term wind speed forecasting that Copula functions are mixed based on time-varying provided in an embodiment of the present invention
Structural representation.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation is described, it is clear that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.Based on this
Embodiment in invention, those of ordinary skill in the art are not under the premise of creative work is made, and what is obtained is every other
Embodiment, belongs to the scope of the present invention.
In order that those skilled in the art more fully understand the present invention program, with reference to the accompanying drawings and detailed description
The present invention is described in further detail.
Next, a kind of short-term wind that Copula functions are mixed based on time-varying that the embodiment of the present invention is provided is discussed in detail
The method of speed prediction.Fig. 1 is a kind of short-term wind speed forecasting that Copula functions are mixed based on time-varying provided in an embodiment of the present invention
Method flow chart, this method includes：
S101：According to the actual value and corresponding predicted value of historical wind speed data, time-varying mixing Copula functions are built.
In embodiments of the present invention, the actual value and corresponding predicted value of historical wind speed data can first be got.
In order to build time-varying mixing Copula functions, the actual value and the respective edge of the predicted value can be first calculated
Distribution function.
For example, X={ p can be usedt| t=1 ..., T } represent air speed data actual value set, ptRepresent a certain moment
The actual value of air speed data,The set of the predicted value of air speed data is represented,Represent a certain moment wind speed
The predicted value of data, X marginal distribution function is FX(x), Y marginal distribution function is FY(y)。
According to follow-up calculating demand, it is necessary to which marginal distribution function is converted into equally distributed form, so that according to institute
Uniformly distributed function is stated, time-varying mixing Copula functions are built.
Specifically, it is possible to use experience distribution, the marginal distribution function is converted into uniformly distributed function.
For example, the form after its conversion is as follows：
Wherein, utWithIt is being uniformly distributed in interval [0,1].
The characteristics of having different due to different Copula functions, and wind speed time series has time-varying characteristics, so
The Copula functions (time-varying mixture copula function, TVMCF) that are mixed using time-varying set up connection
Close distribution function.
In embodiments of the present invention, binary normal state Copula functions, binary Rotated Gumbel Copula can be selected
Function and binary Symmerised Joe-Clayton Copula functions, the Copula functions mixed as time-varying.
Wherein, binary normal state Copula functions describe symmetrical tail-dependence coefficient, and it is defined as follows：
Wherein, ρ ∈ [- 1,1].
What binary Rotated Gumbel Copula functions were portrayed is the correlation of tail on variable, and it is defined as follows：
C(u,v；α)=exp (- [(- ln (1-u))α+(-ln(1-v))α])1/α
Wherein, α ∈ [1 ,+∞).
Binary Symmerised Joe-Clayton Copula functions can portray asymmetrical correlation (i.e. upper tail and
Lower tail is asymmetric), it is defined as follows：
CSJC(u,v|τU,τL(the C of)=0.5JC(u,v|τU,τL)+CJC(1-u,1-v|τU,τL)+u+v-1),
Wherein, τU、τL∈(0,1)。
For the Copula functions of time-varying, its relevant parameter can time to time change.
Accordingly, the structure of time-varying mixing Copula functions is as follows：
Wherein, Cn() represent respectively three kinds of Gaussian Copula, Rotated Gumbel Copula and
Symmerised Joe-Clayton Copula functions；φnFor the relevant parameter of every kind of Copula functions；ηn(0≤ηn≤ 1) be
The weights of every kind of Copula functions, and
S102：Using expectation maximization-maximum likelihood method, the target ginseng of the time-varying mixing Copula functions is determined
Numerical value.
After time-varying mixing Copula functions are built, the relevant parameter of its every kind of Copula function included and correspondingly
Weights belong to unknown target component, operated for the ease of later use time-varying mixing Copula functions, it is necessary to elder generation
Solve the targeted parameter value in time-varying mixing Copula functions.
In embodiments of the present invention, it is possible to use expectation maximization-maximum likelihood (expectation maximization
Integrating with maximum likelihood, EM-ML) method, determine the time-varying mixing Copula functions
Targeted parameter value.
S103：According to the predicted value and time-varying mixing Copula functions, the condition of forecasting wind speed error is calculated
Probability density function.
Predicated error is used to represent the actual value of air speed data and the difference of predicted value.
In embodiments of the present invention, it is possible to use the time-varying mixes Copula functions, sets up the actual value and described
The joint distribution function of predicted value, its formula is as follows：
FXY(x, y)=CM(FX(x),…,FY(y)) (2)
Wherein, function CMWhat is represented is time-varying mixing Copula functions.
Partial derivative on x and y is asked to formula (2), that is, obtains stochastic variable x and y joint probability density function
(probability density function,PDF)fXY, its formula is as follows：
Wherein, fXAnd f (x)Y(y) be respectively x and y marginal probability density function；cM(FX(x),FY(y)) mixed for time-varying
Copula density functions.
For cM(FX(x),FY(y) it) can be asked and be drawn on x and y partial derivative by formula (1), its concrete form is as follows：
Formula (4) is brought into formula (3) and can obtain joint probability density function.
By taking a wind speed value as an example, in the case where giving the wind speed value, the joint probability density letter is utilized
Number, can calculate the condition C opula density functions of its corresponding wind speed actual value.
For example, in given wind speed valueUnder conditions of, the condition of wind speed actual value can be obtained by formula (3)
Copula density functions：
Probabilistic forecasting is set up using forecasting wind speed error, the condition C opula density functions of wind speed actual value are converted
For in predicted valueUnder conditions of, forecasting wind speed errorConditional probability density function (conditional
Probability density function, CPDF), be：
Wherein, predicated error e span is
S104：The conditional probability density function is converted into discrete conditional probability density function, and to described discrete
Conditional probability density function be integrated area accumulation, obtain the forecasting wind speed confidential interval under default confidence level.
The CPDF calculated in S103 is converted into discrete conditional probability density function, the discrete bar can be obtained
The corresponding probable value of various discrete value of part probability density function；The probable value is added up successively, got corresponding
Accumulated value；According to the accumulated value and the default confidence level, corresponding higher limit and lower limit are obtained；According to the upper limit
Value and the lower limit, determine forecasting wind speed confidential interval.
In embodiments of the present invention, significance can be represented with α, default confidence level is 1- α accordingly, calculated
, can be with during upper lower limit valueWithIt is used as judgment condition.
For example, writing down the corresponding probable value of various discrete value for Pj(j=1 ..., N), wherein N are the number of discrete point.It is right
PjProceeded by from j=1 it is cumulative, until the accumulated value of probability is equal toWhen obtain corresponding predicated error lower limit eL, to PjFrom j
=1 proceed by it is cumulative, until the accumulated value of probability is equal toWhen obtain the corresponding predicated error upper limit for eU, accordingly,
The forecast interval that wind speed can be obtained is
From above-mentioned introduction, time-varying mixing Copula functions parameter include every kind of Copula functions parameter and
Corresponding weight coefficient, for the parameter of every kind of Copula functions, can use Maximum Likelihood Estimation Method (maximum respectively
Likelihood, ML) estimated；And for the weight coefficient of every kind of Copula functions, expectation maximization method can be used
(expectation maximization, EM) is determined.
Next the detailed process of parameter and weight coefficient respectively to solving every kind of Copula functions is deployed to introduce.
The parameter of every kind of Copula functions is estimated using ML, comprised the following steps that：
Assuming that sample wind speed actual value X and wind speed value Y have marginal distribution function FX(x；θ1) and FY(y；θ2), side
Edge density function is respectively fX(x；θ1) and fY(y；θ2), wherein θ1、θ2For marginal distribution function and the unknown parameter of density function.
Make u=FX(x；θ1), v=FY(y；θ2), then the distribution function of the Copula functions selected is C (u, v；χ), its is right
In density function beWherein χ is the unknown parameter of Copula functions.Then wind speed actual value X and
Wind speed value Y joint distribution function is：
H(x,y；θ1,θ2, χ) and=C [FX(x；θ1),FY(y；θ2)；χ]
Then corresponding joint probability density function is：
The likelihood function that can obtain sample is：
The both sides of peer-to-peer are taken the logarithm, and can obtain log-likelihood function：
By solving the maximum of points of log-likelihood function, then the unknown parameter of edge distribution and Copula functions can be tried to achieve
Maximum likelihood estimator：
The weight coefficient of every kind of Copula functions is determined using EM, it is comprised the following steps that：
Assuming that an observation ti=(xi,yi) come from some the Copula function c mixed in Copula functionsj, draw
Enter hiding stochastic variable zi=(zi1,zi2,zi3), ziJ-th of element be 1 when, represent sample from jth (j=1,2,3) it is individual
Copula functions.
Introduce hidden variable ziAfterwards, the observation sample is represented by si=(ti,zi), make θ=(θ1,θ2, χ), η=(η1,η2,
η3), λ=(η, θ), then observation sample siConditional probability it is as follows：
Then all observation sample s conditional probability expression formula is：
, can be by asking for the logarithm of above-mentioned condition probability expression seemingly in order to determine the suitable parameters of n observation sample
Right function expectation, i.e.,
Wherein, k represents iterations.
The maximum of conditional expectation can be obtained by above formula, so that it is determined that the estimates of parameters of+1 iteration of kth：
λ(k+1)=argmaxE (lnP (S | λ(k)))
By iteration, it may finally determine to mix the weight coefficient of Copula functions.
Actual value and corresponding predicted value it can be seen from above-mentioned technical proposal according to historical wind speed data, during structure
Become mixing Copula functions；The parameter value of every kind of Copula functions that is included in time-varying mixing Copula functions and corresponding
Weight coefficient belongs to unknown target component, it is possible to use expectation maximization-maximum likelihood method, determines the time-varying mixing
The targeted parameter value of Copula functions.In the case of given wind speed value, Copula functions are mixed using the time-varying, can
To calculate the conditional probability density function of forecasting wind speed error；The conditional probability density function is converted into discrete condition
Probability density function, and area accumulation is integrated to the discrete conditional probability density function, it may finally obtain pre-
If the forecasting wind speed confidential interval under confidence level.It can be seen that, by above-mentioned technical proposal, it can be obtained according to given confidence level
Interval to corresponding wind speed probabilistic forecasting is forecasting wind speed confidential interval, realizes the probabilistic forecasting to wind speed, effectively
The unascertained information for embodying wind speed.
Fig. 2 is a kind of device for the short-term wind speed forecasting that Copula functions are mixed based on time-varying provided in an embodiment of the present invention
Structural representation, including construction unit 21, determining unit 22, computing unit 23 and obtain unit 24,
The construction unit 21, for the actual value and corresponding predicted value according to historical wind speed data, builds time-varying and mixes
Close Copula functions.
The determining unit 22, for utilizing expectation maximization-maximum likelihood method, determines the time-varying mixing
The targeted parameter value of Copula functions；The targeted parameter value includes the parameter value and corresponding weights of every kind of Copula functions
Coefficient.
The computing unit 23, for according to the predicted value and time-varying mixing Copula functions, calculating wind speed
The conditional probability density function of predicated error.
It is described to obtain unit 24, for the conditional probability density function to be converted into discrete conditional probability density letter
Number, and area accumulation is integrated to the discrete conditional probability density function, obtain the wind speed under default confidence level
Prediction confidence intervals.
Optionally, the construction unit includes obtaining subelement, computation subunit, transforming subunit and sets up subelement,
The acquisition subelement, actual value and corresponding predicted value for obtaining historical wind speed data；
The computation subunit, for calculating the actual value and the respective marginal distribution function of the predicted value；
The transforming subunit, for utilizing experience distribution, the marginal distribution function is converted into and is uniformly distributed letter
Number；
It is described to set up subelement, for according to the uniformly distributed function, building time-varying mixing Copula functions.
Optionally, the determining unit determines every kind of Copula functions specifically for utilizing maximum likelihood estimate
Parameter value；Using expectation maximization method, the weight coefficient of every kind of Copula functions is determined.
Optionally, the computing unit includes setting up subelement, obtaining subelement, computation subunit and transforming subunit,
It is described to set up subelement, for mixing Copula functions using the time-varying, set up the actual value and described pre-
The joint distribution function of measured value；
It is described to obtain subelement, for seeking local derviation to the joint distribution function, obtain corresponding joint probability density letter
Number；
The computation subunit, for according to the second predicted value and the joint probability density function, calculating second real
The condition C opula density functions of actual value；
The transforming subunit, for according to predicated error, by the condition C opula density functions of second actual value
It is converted into the conditional probability density function of the predicated error；The predicated error is true according to the actual value and the predicted value
It is fixed.
Optionally, the unit that obtains includes obtaining subelement, obtains subelement and determination subelement,
The acquisition subelement, the various discrete value for obtaining the discrete conditional probability density function is corresponding general
Rate value；
The acquisition subelement is additionally operable to add up to the probable value successively, gets corresponding accumulated value；
It is described to obtain subelement, for according to the accumulated value and the default confidence level, obtaining corresponding higher limit
And lower limit；
The determination subelement, for according to the higher limit and the lower limit, determining forecasting wind speed confidential interval
The explanation of feature may refer to the related description of embodiment corresponding to Fig. 1 in embodiment corresponding to Fig. 2, here no longer
Repeat one by one.
Actual value and corresponding predicted value it can be seen from above-mentioned technical proposal according to historical wind speed data, during structure
Become mixing Copula functions；The parameter value of every kind of Copula functions that is included in time-varying mixing Copula functions and corresponding
Weight coefficient belongs to unknown target component, it is possible to use expectation maximization-maximum likelihood method, determines the time-varying mixing
The targeted parameter value of Copula functions.In the case of given wind speed value, Copula functions are mixed using the time-varying, can
To calculate the conditional probability density function of forecasting wind speed error；The conditional probability density function is converted into discrete condition
Probability density function, and area accumulation is integrated to the discrete conditional probability density function, it may finally obtain pre-
If the forecasting wind speed confidential interval under confidence level.It can be seen that, by above-mentioned technical proposal, it can be obtained according to given confidence level
Interval to corresponding wind speed probabilistic forecasting is forecasting wind speed confidential interval, realizes the probabilistic forecasting to wind speed, effectively
The unascertained information for embodying wind speed.
A kind of short-term wind speed forecasting that Copula functions are mixed based on time-varying provided above the embodiment of the present invention
Method and apparatus is described in detail.The embodiment of each in specification is described by the way of progressive, each embodiment emphasis
What is illustrated is all the difference with other embodiment, between each embodiment identical similar portion mutually referring to.For
For device disclosed in embodiment, because it is corresponded to the method disclosed in Example, so description is fairly simple, correlation
Place is referring to method part illustration.It should be pointed out that for those skilled in the art, not departing from this hair
On the premise of bright principle, some improvement and modification can also be carried out to the present invention, these are improved and modification also falls into power of the present invention
In the protection domain that profit is required.
Professional further appreciates that, with reference to the unit of each example of the embodiments described herein description
And algorithm steps, can be realized with electronic hardware, computer software or the combination of the two, in order to clearly demonstrate hardware and
The interchangeability of software, generally describes the composition and step of each example according to function in the above description.These
Function is performed with hardware or software mode actually, depending on the application-specific and design constraint of technical scheme.Specialty
Technical staff can realize described function to each specific application using distinct methods, but this realization should not
Think beyond the scope of this invention.
Directly it can be held with reference to the step of the method or algorithm that the embodiments described herein is described with hardware, processor
Capable software module, or the two combination are implemented.Software module can be placed in random access memory (RAM), internal memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
Claims (10)
1. a kind of method for the short-term wind speed forecasting that Copula functions are mixed based on time-varying, it is characterised in that including：
According to the actual value and corresponding predicted value of historical wind speed data, time-varying mixing Copula functions are built；
Using expectation maximization-maximum likelihood method, determine that the time-varying mixes the targeted parameter value of Copula functions；It is described
Targeted parameter value includes the parameter value and corresponding weight coefficient of every kind of Copula functions；
According to the predicted value and time-varying mixing Copula functions, the conditional probability density letter of forecasting wind speed error is calculated
Number；
The conditional probability density function is converted into discrete conditional probability density function, and to the discrete conditional probability
Density function is integrated area accumulation, obtains the forecasting wind speed confidential interval under default confidence level.
2. according to the method described in claim 1, it is characterised in that the actual value according to air speed data and corresponding prediction
Value, building time-varying mixing Copula functions includes：
Obtain the actual value and corresponding predicted value of historical wind speed data；
Calculate the actual value and the respective marginal distribution function of the predicted value；
Using experience distribution, the marginal distribution function is converted into uniformly distributed function；
According to the uniformly distributed function, time-varying mixing Copula functions are built.
3. method according to claim 2, it is characterised in that the utilization expectation maximization-maximum likelihood method, it is determined that
Going out the targeted parameter value of the time-varying mixing Copula functions includes：
Using maximum likelihood estimate, the parameter value of every kind of Copula functions is determined；
Using expectation maximization method, the weight coefficient of every kind of Copula functions is determined.
4. method according to claim 3, it is characterised in that described to be mixed according to the predicted value and the time-varying
Copula functions, calculating the conditional probability density function of forecasting wind speed error includes：
Copula functions are mixed using the time-varying, the joint distribution function of the actual value and the predicted value is set up；
Local derviation is asked to the joint distribution function, corresponding joint probability density function is obtained；
According to the second predicted value and the joint probability density function, the condition C opula density letters of the second actual value are calculated
Number；
According to predicated error, the condition C opula density functions of second actual value are converted into the condition of the predicated error
Probability density function；The predicated error is determined according to the actual value and the predicted value.
5. method according to claim 4, it is characterised in that described to be carried out to the discrete conditional probability density function
Integral area adds up, and the forecasting wind speed confidential interval obtained under default confidence level includes：
Obtain the corresponding probable value of various discrete value of the discrete conditional probability density function；
The probable value is added up successively, corresponding accumulated value is got；
According to the accumulated value and the default confidence level, corresponding higher limit and lower limit are obtained；
According to the higher limit and the lower limit, forecasting wind speed confidential interval is determined.
6. it is a kind of based on time-varying mix Copula functions short-term wind speed forecasting device, it is characterised in that including construction unit,
Determining unit, computing unit and unit is obtained,
The construction unit, for the actual value and corresponding predicted value according to historical wind speed data, builds time-varying mixing
Copula functions；
The determining unit, for utilizing expectation maximization-maximum likelihood method, determines the time-varying mixing Copula functions
Targeted parameter value；The targeted parameter value includes the parameter value and corresponding weight coefficient of every kind of Copula functions；
The computing unit, for according to the predicted value and time-varying mixing Copula functions, calculating forecasting wind speed mistake
The conditional probability density function of difference；
It is described to obtain unit, for the conditional probability density function to be converted into discrete conditional probability density function, and it is right
The discrete conditional probability density function is integrated area accumulation, obtains the forecasting wind speed confidence under default confidence level
It is interval.
7. device according to claim 6, it is characterised in that the construction unit includes obtaining subelement, calculates son list
Member, transforming subunit and subelement is set up,
The acquisition subelement, actual value and corresponding predicted value for obtaining historical wind speed data；
The computation subunit, for calculating the actual value and the respective marginal distribution function of the predicted value；
The transforming subunit, for utilizing experience distribution, uniformly distributed function is converted into by the marginal distribution function；
It is described to set up subelement, for according to the uniformly distributed function, building time-varying mixing Copula functions.
8. device according to claim 7, it is characterised in that the determining unit is specifically for utilizing maximal possibility estimation
Method, determines the parameter value of every kind of Copula functions；Using expectation maximization method, the weights system of every kind of Copula functions is determined
Number.
9. device according to claim 8, it is characterised in that the computing unit includes setting up subelement, obtains sub single
Member, computation subunit and transforming subunit,
It is described to set up subelement, for mixing Copula functions using the time-varying, set up the actual value and the predicted value
Joint distribution function；
It is described to obtain subelement, for seeking local derviation to the joint distribution function, obtain corresponding joint probability density function；
The computation subunit, for according to the second predicted value and the joint probability density function, calculating the second actual value
Condition C opula density functions；
The transforming subunit, for according to predicated error, the condition C opula density functions of second actual value to be converted
For the conditional probability density function of the predicated error；The predicated error is determined according to the actual value and the predicted value.
10. device according to claim 9, it is characterised in that the unit that obtains includes obtaining subelement, obtains sub single
Member and determination subelement,
The acquisition subelement, the corresponding probability of various discrete value for obtaining the discrete conditional probability density function
Value；
The acquisition subelement is additionally operable to add up to the probable value successively, gets corresponding accumulated value；
It is described to obtain subelement, for according to the accumulated value and the default confidence level, obtaining corresponding higher limit with
Limit value；
The determination subelement, for according to the higher limit and the lower limit, determining forecasting wind speed confidential interval.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710485792.XA CN107292439A (en) | 2017-06-23 | 2017-06-23 | A kind of method and apparatus for the short-term wind speed forecasting that Copula functions are mixed based on time-varying |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710485792.XA CN107292439A (en) | 2017-06-23 | 2017-06-23 | A kind of method and apparatus for the short-term wind speed forecasting that Copula functions are mixed based on time-varying |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107292439A true CN107292439A (en) | 2017-10-24 |
Family
ID=60097946
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710485792.XA Pending CN107292439A (en) | 2017-06-23 | 2017-06-23 | A kind of method and apparatus for the short-term wind speed forecasting that Copula functions are mixed based on time-varying |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107292439A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110084403A (en) * | 2019-03-27 | 2019-08-02 | 山东信通电子股份有限公司 | The multi-period joint probability density prediction technique and system of aerial condutor current-carrying capacity |
| CN111159900A (en) * | 2019-12-30 | 2020-05-15 | 国核电力规划设计研究院有限公司 | Method and device for predicting wind speed of fan |
| CN112183866A (en) * | 2020-09-29 | 2021-01-05 | 中南大学 | Short-term wind speed exceeding probability prediction method and system |
| CN112685915A (en) * | 2021-01-18 | 2021-04-20 | 重庆大学 | Wind power output condition probability distribution modeling method |
| CN113312836A (en) * | 2021-05-10 | 2021-08-27 | 国核电力规划设计研究院有限公司 | Short-term wind speed prediction method |
| CN113487097A (en) * | 2021-07-12 | 2021-10-08 | 国网新疆电力有限公司电力科学研究院 | Ultra-short-term wind speed interval prediction method based on data fusion |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102945223A (en) * | 2012-11-21 | 2013-02-27 | 华中科技大学 | Method for constructing joint probability distribution function of output of a plurality of wind power plants |
| CN103020423A (en) * | 2012-11-21 | 2013-04-03 | 华中科技大学 | Copula-function-based method for acquiring relevant characteristic of wind power plant capacity |
| CN104915788A (en) * | 2015-06-24 | 2015-09-16 | 华南理工大学 | Method for electric power system dynamic economic dispatching with correlation of multiple wind power plants being considered |
| CN105095674A (en) * | 2015-09-07 | 2015-11-25 | 国网天津市电力公司 | Distributed fan output correlation scenarios analysis method |
| CN105426998A (en) * | 2015-11-19 | 2016-03-23 | 广西大学 | Method for predicting wind power interval based on multiple conditions |
| CN106058920A (en) * | 2016-06-06 | 2016-10-26 | 南京理工大学 | Method for optimizing spinning reserve capacity of wind farm based on Mixed-Copula function |
| CN106485362A (en) * | 2016-10-18 | 2017-03-08 | 江苏省电力试验研究院有限公司 | A kind of power generation dispatching method based on higher-dimension wind-powered electricity generation forecast error model and dimensionality reduction technology |
-
2017
- 2017-06-23 CN CN201710485792.XA patent/CN107292439A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102945223A (en) * | 2012-11-21 | 2013-02-27 | 华中科技大学 | Method for constructing joint probability distribution function of output of a plurality of wind power plants |
| CN103020423A (en) * | 2012-11-21 | 2013-04-03 | 华中科技大学 | Copula-function-based method for acquiring relevant characteristic of wind power plant capacity |
| CN104915788A (en) * | 2015-06-24 | 2015-09-16 | 华南理工大学 | Method for electric power system dynamic economic dispatching with correlation of multiple wind power plants being considered |
| CN105095674A (en) * | 2015-09-07 | 2015-11-25 | 国网天津市电力公司 | Distributed fan output correlation scenarios analysis method |
| CN105426998A (en) * | 2015-11-19 | 2016-03-23 | 广西大学 | Method for predicting wind power interval based on multiple conditions |
| CN106058920A (en) * | 2016-06-06 | 2016-10-26 | 南京理工大学 | Method for optimizing spinning reserve capacity of wind farm based on Mixed-Copula function |
| CN106485362A (en) * | 2016-10-18 | 2017-03-08 | 江苏省电力试验研究院有限公司 | A kind of power generation dispatching method based on higher-dimension wind-powered electricity generation forecast error model and dimensionality reduction technology |
Non-Patent Citations (3)
| Title |
|---|
| XIANGANG PENG ET AL.: "A novel probabilistic wind speed forecasting based on combination of", 《ENERGY CONVERSION AND MANAGEMENT》 * |
| 兰飞等: "基于条件Copula 函数的风电功率区间预测", 《中国电机工程学报》 * |
| 韩晓庆: "Copula函数的理论及其应用", 《中国优秀硕士学位论文全文数据库 基础科学辑》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110084403A (en) * | 2019-03-27 | 2019-08-02 | 山东信通电子股份有限公司 | The multi-period joint probability density prediction technique and system of aerial condutor current-carrying capacity |
| CN111159900A (en) * | 2019-12-30 | 2020-05-15 | 国核电力规划设计研究院有限公司 | Method and device for predicting wind speed of fan |
| CN111159900B (en) * | 2019-12-30 | 2024-02-09 | 国核电力规划设计研究院有限公司 | Method and device for predicting wind speed of fan |
| CN112183866A (en) * | 2020-09-29 | 2021-01-05 | 中南大学 | Short-term wind speed exceeding probability prediction method and system |
| CN112685915A (en) * | 2021-01-18 | 2021-04-20 | 重庆大学 | Wind power output condition probability distribution modeling method |
| CN112685915B (en) * | 2021-01-18 | 2023-06-30 | 重庆大学 | Wind power output condition probability distribution modeling method |
| CN113312836A (en) * | 2021-05-10 | 2021-08-27 | 国核电力规划设计研究院有限公司 | Short-term wind speed prediction method |
| CN113487097A (en) * | 2021-07-12 | 2021-10-08 | 国网新疆电力有限公司电力科学研究院 | Ultra-short-term wind speed interval prediction method based on data fusion |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107292439A (en) | A kind of method and apparatus for the short-term wind speed forecasting that Copula functions are mixed based on time-varying | |
| Mehrotra et al. | Conditional resampling of hydrologic time series using multiple predictor variables: A K-nearest neighbour approach | |
| Tencaliec et al. | Flexible semiparametric generalized Pareto modeling of the entire range of rainfall amount | |
| Daescu et al. | Adaptive observations in the context of 4D-Var data assimilation | |
| Kitiashvili et al. | Application of data assimilation method for predicting solar cycles | |
| CN112686464A (en) | Short-term wind power prediction method and device | |
| CN102968573A (en) | Online lithium ion battery residual life predicting method based on relevance vector regression | |
| CN107274030B (en) | Runoff Forecast method and system based on hydrology variable year border and monthly variation characteristic | |
| CN103488869A (en) | Wind power generation short-term load forecast method of least squares support vector machine | |
| CN103942461A (en) | Water quality parameter prediction method based on online sequential extreme learning machine | |
| CN110910004A (en) | Reservoir dispatching rule extraction method and system with multiple uncertainties | |
| CN104573876A (en) | Wind power plant short-period wind speed prediction method based on time sequence long memory model | |
| CN109787251B (en) | Cluster temperature control load aggregation model, system parameter identification and reverse control method | |
| Wang et al. | Monthly streamflow simulation for the headwater catchment of the Yellow River basin with a hybrid statistical‐dynamical model | |
| Loucks et al. | An introduction to probability, statistics, and uncertainty | |
| CN103839113B (en) | Microscopic simulation method based on house selecting models of house-renting selectors and house selecting models of house-purchasing selectors | |
| Zhou et al. | A robust segmented mixed effect regression model for baseline electricity consumption forecasting | |
| CN103885867A (en) | Online evaluation method of performance of analog circuit | |
| CN105449667A (en) | Prediction method for reliability of power generation system and power transmission system | |
| CN105939026B (en) | Wind power undulate quantity probability Distribution Model method for building up based on mixing Laplace distributions | |
| Parviz et al. | Development of precipitation forecast model based on artificial intelligence and subseasonal clustering | |
| Darren Graham et al. | Lognormal distribution provides an optimum representation of the concrete delivery and placement process | |
| CN100420209C (en) | Credible network simulation system of automatic conceptual contrast | |
| CN110069708A (en) | A kind of promotion effect predictor method, device, medium and equipment promoted across medium | |
| Sun et al. | An improved and efficient estimation method for varying-coefficient model with missing covariates |
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: 20171024 |
|
| RJ01 | Rejection of invention patent application after publication |