CN107067099A - Wind power probability forecasting method and device - Google Patents
Wind power probability forecasting method and device Download PDFInfo
- Publication number
- CN107067099A CN107067099A CN201710055611.XA CN201710055611A CN107067099A CN 107067099 A CN107067099 A CN 107067099A CN 201710055611 A CN201710055611 A CN 201710055611A CN 107067099 A CN107067099 A CN 107067099A
- Authority
- CN
- China
- Prior art keywords
- wind power
- condition
- wind
- error
- power plant
- 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
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
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- 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 wind power probability forecasting method and device, methods described includes：According to history power output and historical forecast power, the statistical nature of wind power plant predicated error is obtained；The fluctuations in wind speed amount predicted the outcome according to the NWP of historical forecast power and wind power plant, obtains the condition complete or collected works of wind-powered electricity generation probabilistic forecasting；By K means clustering algorithms, condition complete or collected works are divided into some condition subsets；Error set formation condition experience distribution to being under each condition subset, and examine whether its numerical characteristic overlaps with numerical characteristic in the statistical nature of wind power plant predicated error；Clustered again by K means clustering algorithms if overlapping；And the wind power prediction result according to each moment and the distribution of condition experience, obtain wind power probabilistic forecasting result.The invention further relates to a kind of prediction meanss.The wind power probability forecasting method that the present invention is provided can provide error distribution function to otherness, with higher prediction accuracy.
Description
Technical field
The present invention relates to a kind of wind power probability forecasting method, more particularly to a kind of wind power probability forecasting method and
Device.
Background technology
The randomness of wind-resources, fluctuation, the uncontrollability of uncertainty and wind power output are steady to the safety of power system
Fixed operation brings larger puzzlement.Wind power prediction technology turns into indispensable technology.
However, in actual motion, the predicated error in wind power prediction result can not be avoided, traditional wind power
The general point prediction using wind power of prediction, so as to can not provide wind power probabilistic information.Conventional wind power
There is unpredictable predicated error in prediction, cause forecasting inaccuracy really, the security and stability analysis for influenceing power system, from
And influence the validity of operational decisions result.
The content of the invention
In summary, it is necessory to propose a kind of method and device that accurately can be predicted to wind power probability.
A kind of wind power probability forecasting method, comprises the following steps：
According to history power output and historical forecast power, the statistical nature of wind power plant predicated error is obtained；
The fluctuations in wind speed amount predicted the outcome according to the NWP of historical forecast power and wind power plant, obtains wind-powered electricity generation probabilistic forecasting
Condition complete or collected works；
By K-means clustering algorithms, condition complete or collected works are divided into some condition subsets；
Error set formation condition experience distribution to being under each condition subset, and examine its numerical characteristic and wind-powered electricity generation
Whether numerical characteristic overlaps in the statistical nature of field prediction error；Gathered again by K-means clustering algorithms if overlapping
Class；And
It is distributed according to the wind power prediction result at each moment and condition experience, obtains wind power probabilistic forecasting knot
Really.
In one of the embodiments, the statistical nature of the wind power plant predicated error includes the equal of wind power plant predicated error
Value E (X), variance Var (X), degree of bias Skew (X) and kurtosis Kurtosis (X)；
In one of the embodiments, it is described according to history power output and historical forecast power, obtain wind power plant pre-
The step of statistical nature for surveying error, includes：
Obtain the history power output of wind power plant within a predetermined period of time；
Obtain the historical forecast power of wind power plant within a predetermined period of time；
Calculate the error vector in predetermined amount of time；And
Each component in error vector is obtained, wind power plant predicated error sample value is constituted, calculates wind power plant predicated error
Statistical nature.
In one of the embodiments, the wind speed ripple that the NWP according to historical forecast power and wind power plant predicts the outcome
The step of momentum, condition complete or collected works for obtaining wind-powered electricity generation probabilistic forecasting, includes：
According to the historical forecast power of wind power plant in the predetermined amount of time of acquisition, obtain in moment t, wind power plant NWP is pre-
Survey the fluctuations in wind speed amount of result；
Obtain the fluctuations in wind speed duration set that the fluctuations in wind speed amount at the moment from t-s to t-1 is constituted；And
The Descartes's collection for being mutually combined and being constituted by fluctuations in wind speed amount and the set is obtained, condition complete or collected works are used as.
In one of the embodiments, it is described by K-means clustering algorithms, condition complete or collected works are divided into some condition
The step of collection, includes：
For condition complete or collected works, using K-means clustering algorithms, m clusters are classified as, that is, are classified as m individual mutually disjoint
Condition subset, m numerical value is evenly distributed for the element number in 4~10, and each condition subset.
In one of the embodiments, the described pair of error set formation condition experience being under each condition subset is divided
Cloth, and include the step of examine its numerical characteristic and whether numerical characteristic overlaps in the statistical nature of wind power plant predicated error：
To each condition subset Ci, obtain error sample set E under this conditioni, obtain wind-powered electricity generation work(at different conditions
The experience distribution PDF of rate predicated errori, and be distributed to describe in condition C with the experience of wind power prediction erroriUnder error
Stochastic variable ei:
ei~PDFi；
The condition experience distribution of m error of one by one inspection, if special with numeral in the statistical nature of wind power plant predicated error
The error empirical distribution function levied is overlapped.
It is in one of the embodiments, described to be distributed according to the wind power prediction result and condition experience at each moment,
The step of obtaining wind power probabilistic forecasting result includes：
The wind power prediction result at moment t+k is obtained in tAnd moment t+k prediction of wind speed undulate quantity
fluct+k；
If real number pairIn condition subset Cj, select the error experience distribution PDF of the condition subsetj,
And by error stochastic variable ejIt is used as the error stochastic variable at the t+k moment；
With reference to t+k wind power prediction resultObtain final probabilistic forecasting result：
Wherein,It is the result of wind power probabilistic forecasting, is stochastic variable, for describes the uncertain of wind power
Information.
A kind of wind power probabilistic forecasting device, described device includes：
Feature calculation module, for according to history power output and historical forecast power, obtaining wind power plant predicated error
Statistical nature；
Condition computing module, for the fluctuations in wind speed amount predicted the outcome according to the NWP of historical forecast power and wind power plant, is obtained
Obtain the condition complete or collected works of wind-powered electricity generation probabilistic forecasting；
Sort module, for by K-means clustering algorithms, condition complete or collected works to be divided into some condition subsets；
Cluster module, for being distributed to the error set formation condition experience being under each condition subset, and examines it
Whether numerical characteristic overlaps with numerical characteristic in the statistical nature of wind power plant predicated error；Clustered if overlapping by K-means
Algorithm is clustered again；
Prediction module, is distributed for the wind power prediction result according to each moment and condition experience, obtains wind-powered electricity generation work(
Rate probabilistic forecasting result.
The feature calculation module includes：
First acquisition unit, for obtaining the history power output of wind power plant within a predetermined period of time；
Second acquisition unit, the historical forecast power for obtaining wind power plant within a predetermined period of time；
First computing unit, for calculating the error in predetermined amount of time between history power output and historical forecast power
Vector；
Second computing unit, for obtaining each component in error vector, constitutes wind power plant predicated error sample value, calculates
The statistical nature of wind power plant predicated error.
The condition computing module includes：
Undulate quantity acquiring unit, for the historical forecast power of wind power plant in the predetermined amount of time according to acquisition, is obtained
During moment t, the fluctuations in wind speed amount that wind power plant NWP predicts the outcome；
Duration set acquiring unit is fluctuated, for obtaining the fluctuations in wind speed amount that the fluctuations in wind speed amount at the moment from t-s to t-1 is constituted
Set；
Condition complete or collected works' acquiring unit, the Descartes constituted is mutually combined for obtaining by fluctuations in wind speed amount and the set
Collection, is used as condition complete or collected works.
The wind power probabilistic forecasting technology that the present invention is provided, using conditional probability distribution and K-means clustering algorithms as base
Plinth, can provide error distribution function to otherness, with higher prediction accuracy.
Brief description of the drawings
The FB(flow block) for the wind power probabilistic forecasting that Fig. 1 provides for present example.
Embodiment
Further stated in detail below according to Figure of description and in conjunction with specific embodiments to technical scheme.
In large-scale wind electricity cluster, the power output of adjacent multiple wind power plants has stronger correlation, therefore can combine
Wind power plant i wind speed measured value, and adjacent multiple strong correlation Power Output for Wind Power Field measured values carry out the exception of the wind power plant
Data point calibration, by wind power plant i moment t exceptional data pointIt is corrected to what is estimated according to wind speed and spatial coherence
Corrected power estimate
Realize that wind power plant i abnormal data point calibration needs to solve two key issues：First, certainty equivalence surveys wind speed and isWhen, the zone of reasonableness of power output value.When known to wind power plant equivalent power curve, the reasonable model of power output value
Enclose as wind speedThe span determined with equivalent power curve bound intersection point.Second, determine wind power plant i current powers
Value, can be based on the conditional probability distribution of strong correlation wind power plant (for example coefficient correlation is more than 0.5), carry out the sampling of condition recurrence and come
It is determined that.Here only consider the reason for strong correlation wind power plant to be that conditional probability distribution dimension can be reduced, it is to avoid dimension calamity occur, and
And for wind farm data correction, it is considered to strong correlation wind power plant can meet correction demand.
Referring to Fig. 1, the wind power probabilistic forecasting side based on clustering algorithm and error analysis that present example is provided
Method, comprises the following steps：
Step S10, obtains the statistical nature of wind power plant predicated error.
Specifically, in one of the embodiments, the acquisition of the statistical nature of wind power plant predicated error includes：
Step S11, obtains the history power output of wind power plant within a predetermined period of time, is：
Wherein, t is current time, and s is the historical data duration reviewed forward.
Step S12, obtains the historical forecast power of wind power plant within a predetermined period of time, is：
Step S13, calculates the error vector between the history power output and historical forecast power in predetermined amount of time,
For：
Step S14, obtains each component in error vector, constitutes wind power plant predicated error sample value, calculate wind power plant pre-
Survey the statistical nature of error.
Specifically, the statistical nature of wind power plant predicated error may include average E (X), the variance Var of wind power plant predicated error
(X), degree of bias Skew (X) and kurtosis Kurtosis (X).The X can be made to beIn the wind power plant predicated error sample that constitutes of each component
Collection, calculates the statistical nature of wind power plant predicated error：Average E (X), variance Var (X), degree of bias Skew (X) and kurtosis Kurtosis
(X)。
Step S20, according to historical forecast power and the NWP of wind power plant (numerical weather forecast, numerical weather
Prediection the fluctuations in wind speed amount) predicted the outcome, obtains the condition complete or collected works of wind-powered electricity generation probabilistic forecasting.
Specifically, in one of the embodiments, the condition complete or collected works of wind-powered electricity generation probabilistic forecasting and the acquisition of scatter diagram include
Following steps：
Step S21, according to the historical forecast power of wind power plant in the predetermined amount of time of acquisition：
Obtain in moment t, the fluctuations in wind speed amount that wind power plant NWP predicts the outcome is：
Step S22, selects the fluctuations in wind speed amount fluc at the moment from t-s to t-1t-s,…,fluct-1The fluctuations in wind speed of composition
Quantity set is combined into Fluct,s。
Step S23, obtain byAnd Fluct,sBoth are mutually combined Descartes's collection of composition, as condition complete or collected works C, and
Corresponding scatter diagram can be formed.
Condition complete or collected works, by K-means clustering algorithms, are divided into some condition subsets by step S30.
Condition complete or collected works C can be classified by K-means clustering algorithms, formation includes C1,C2,...,CmM bar
Part subset.
K-means algorithms are the clustering algorithms based on distance, using evaluation index of the distance as similitude, that is, think two
The distance of individual object is nearer, and its similarity is bigger.
Specifically, for condition complete or collected works C, using K-means clustering algorithms, being classified as m clusters, that is, it is classified as m mutually
Disjoint condition subset：
M numerical value selection can be 4~10, selected according to different situations, should ensure that the element in each condition subset
Number is more average, i.e. the number of element in each condition subset is essentially identical.
Step S40, to being in each condition subset CiUnder error set formation condition experience distribution, and examine its numeral
Whether height is overlapped feature with numerical characteristic in the statistical nature of wind power plant predicated error；The return to step S30 if highly overlapping,
Re-start cluster.
Specifically, in one of the embodiments, to each condition subset Ci, error sample under this condition can be obtained
Collect Ei, the experience distribution PDF of wind power prediction error at different conditions can be obtainedi, described with this distribution in bar
Part CiUnder error stochastic variable ei:
ei~PDFi；
Whether the condition experience distribution of m error of one by one inspection is highly heavy with the error empirical distribution function in step S10
Close.So-called height is overlapped, and is that both numerical characteristic (average, variance, the degree of bias, kurtosis) differences are less than predetermined threshold value, this is preset
Threshold value can be selected as needed, and predetermined threshold value is smaller, then numerical characteristic is more identical.
If highly overlapping, the condition difference information that error can not be distributed by illustrating this cluster is effectively peeled off.Need to return
Step S30, the numerical value for changing m re-starts cluster, until the distribution of none of condition experience and wind power plant predicated error
Statistical nature in experience distribution height overlap.
Step S50, with reference to the wind power prediction result formation wind power probabilistic forecasting result at each moment.
Specifically, in one of the embodiments, the acquisition of wind power probabilistic forecasting result includes：
Step S51, the wind power prediction result at moment t+k is obtained in tAnd moment t+k prediction of wind speed
Undulate quantity fluct+k；
Step S52, if this real number pairIn condition subset Cj, select the error of this condition subset
Experience is distributed PDFj, by error stochastic variable ejIt is used as the error stochastic variable at the t+k moment；
Step S53, with reference to t+k wind power prediction resultForm final probabilistic forecasting result：
Wherein,It is the result of wind power probabilistic forecasting, is a stochastic variable, wind power can be described not
Determine information.
Wind power probability forecasting method provided in an embodiment of the present invention, is calculated with conditional probability distribution and K-means clusters
Based on method, error distribution function can be provided to otherness, with more preferable prediction effect.The wind power probabilistic forecasting side
Method is modeled and analyzed by the predicated error to wind power, can be to the possible fluctuation range of wind power and probability point
The description of overall picture is furnished with, so as to provide reliable, detailed information for wind power prediction result.
Based on same inventive concept, the embodiment of the present invention additionally provides a kind of wind power probabilistic forecasting device, due to this
The principle that a little equipment solve problem is similar to a kind of wind power probability forecasting method, therefore the implementation of these equipment may refer to
The implementation of method, repeats part and repeats no more.
The wind power probabilistic forecasting device can include：
Feature calculation module, for according to history power output and historical forecast power, obtaining wind power plant predicated error
Statistical nature；
Condition computing module, for the fluctuations in wind speed amount predicted the outcome according to the NWP of historical forecast power and wind power plant, is obtained
Obtain the condition complete or collected works of wind-powered electricity generation probabilistic forecasting；
Sort module, for by K-means clustering algorithms, condition complete or collected works to be divided into some condition subsets；
Cluster module, for being distributed to the error set formation condition experience being under each condition subset, and examines it
Whether numerical characteristic overlaps with numerical characteristic in the statistical nature of wind power plant predicated error；Clustered if overlapping by K-means
Algorithm is clustered again；
Prediction module, is distributed for the wind power prediction result according to each moment and condition experience, obtains wind-powered electricity generation work(
Rate probabilistic forecasting result.
In implementation, the feature calculation module can include：
First acquisition unit, for obtaining the history power output of wind power plant within a predetermined period of time；
Second acquisition unit, the historical forecast power for obtaining wind power plant within a predetermined period of time；
First computing unit, for calculating the error in predetermined amount of time between history power output and historical forecast power
Vector；
Second computing unit, for obtaining each component in error vector, constitutes wind power plant predicated error sample value, calculates
The statistical nature of wind power plant predicated error.
In implementation, the condition computing module can include：
Undulate quantity acquiring unit, for the historical forecast power of wind power plant in the predetermined amount of time according to acquisition, is obtained
During moment t, the fluctuations in wind speed amount that wind power plant NWP predicts the outcome；
Duration set acquiring unit is fluctuated, for obtaining the fluctuations in wind speed amount that the fluctuations in wind speed amount at the moment from t-s to t-1 is constituted
Set；
Condition complete or collected works' acquiring unit, the Descartes constituted is mutually combined for obtaining by fluctuations in wind speed amount and the set
Collection, is used as condition complete or collected works.
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 wind power probability forecasting method, it is characterised in that methods described includes：
According to history power output and historical forecast power, the statistical nature of wind power plant predicated error is obtained；
The fluctuations in wind speed amount predicted the outcome according to the numerical weather forecast NWP of historical forecast power and wind power plant, obtains wind-powered electricity generation general
The condition complete or collected works of rate prediction；
By K-means clustering algorithms, condition complete or collected works are divided into some condition subsets；
Error set formation condition experience distribution to being under each condition subset, and examine its numerical characteristic and wind power plant pre-
Survey whether numerical characteristic in the statistical nature of error overlaps；Clustered again by K-means clustering algorithms if overlapping；With
And
It is distributed according to the wind power prediction result at each moment and condition experience, obtains wind power probabilistic forecasting result.
2. wind power probability forecasting method according to claim 1, it is characterised in that the wind power plant predicated error
Statistical nature includes average E (X), variance Var (X), degree of bias Skew (X) and the kurtosis Kurtosis (X) of wind power plant predicated error.
3. wind power probability forecasting method according to claim 2, it is characterised in that described according to history power output
And historical forecast power, obtain wind power plant predicated error statistical nature the step of include：
Obtain the history power output of wind power plant within a predetermined period of time；
Obtain the historical forecast power of wind power plant within a predetermined period of time；
Calculate the error vector between history power output and historical forecast power in predetermined amount of time；And
Each component in error vector is obtained, wind power plant predicated error sample value is constituted, calculates the statistics of wind power plant predicated error
Feature.
4. wind power probability forecasting method according to claim 1, it is characterised in that described according to historical forecast power
The fluctuations in wind speed amount predicted the outcome with the numerical weather forecast NWP of wind power plant, obtains the step of the condition complete or collected works of wind-powered electricity generation probabilistic forecasting
Suddenly include：
According to the historical forecast power of wind power plant in the predetermined amount of time of acquisition, obtain in moment t, wind power plant NWP prediction knots
The fluctuations in wind speed amount of fruit；
Obtain the fluctuations in wind speed duration set that the fluctuations in wind speed amount at the moment from t-s to t-1 is constituted；And
The Descartes's collection for being mutually combined and being constituted by fluctuations in wind speed amount and the set is obtained, condition complete or collected works are used as.
5. wind power probability forecasting method according to claim 1, it is characterised in that described to be clustered by K-means
Algorithm, the step of condition complete or collected works are divided into some condition subsets includes：
For condition complete or collected works, using K-means clustering algorithms, m clusters are classified as, that is, are classified as m mutually disjoint conditions
Subset, m numerical value is evenly distributed for the element number in 4~10, and each condition subset.
6. wind power probability forecasting method according to claim 5, it is characterised in that described pair is in each condition
Error set formation condition experience distribution under collection, and examine its numerical characteristic and number in the statistical nature of wind power plant predicated error
The step of whether word feature overlaps includes：
To each condition subset Ci, obtain error sample set E under this conditioni, the wind power obtained at different conditions is pre-
Survey the experience distribution PDF of errori, and be distributed to describe in condition C with the experience of wind power prediction erroriUnder error it is random
Variable ei:
ei~PDFi；
The condition experience distribution of m error of one by one inspection, if with numerical characteristic in the statistical nature of wind power plant predicated error
Error empirical distribution function is overlapped.
7. wind power probability forecasting method according to claim 1, it is characterised in that the wind-powered electricity generation according to each moment
Power prediction result and the distribution of condition experience, the step of obtaining wind power probabilistic forecasting result include：
The wind power prediction result at moment t+k is obtained in tAnd moment t+k prediction of wind speed undulate quantity fluct+k；
If real number to (fluct+k) it is in condition subset Cj, select the error experience distribution PDF of the condition subsetj, and will be by mistake
Poor stochastic variable ejIt is used as the error stochastic variable at the t+k moment；
With reference to t+k wind power prediction resultObtain final probabilistic forecasting result：
<mrow>
<msubsup>
<mi>P</mi>
<mrow>
<mi>f</mi>
<mo>,</mo>
<mi>p</mi>
</mrow>
<mrow>
<mi>t</mi>
<mo>+</mo>
<mi>k</mi>
</mrow>
</msubsup>
<mo>=</mo>
<msubsup>
<mi>P</mi>
<mi>f</mi>
<mrow>
<mi>t</mi>
<mo>+</mo>
<mi>k</mi>
</mrow>
</msubsup>
<mo>+</mo>
<msub>
<mi>e</mi>
<mi>j</mi>
</msub>
<mo>;</mo>
</mrow>
Wherein,It is the result of wind power probabilistic forecasting, is stochastic variable, the uncertain information for describing wind power.
8. a kind of wind power probabilistic forecasting device, it is characterised in that described device includes：
Feature calculation module, for according to history power output and historical forecast power, obtaining the system of wind power plant predicated error
Count feature；
Condition computing module, for the wind speed predicted the outcome according to the numerical weather forecast NWP of historical forecast power and wind power plant
Undulate quantity, obtains the condition complete or collected works of wind-powered electricity generation probabilistic forecasting；
Sort module, for by K-means clustering algorithms, condition complete or collected works to be divided into some condition subsets；
Cluster module, for being distributed to the error set formation condition experience being under each condition subset, and examines its numeral
Whether feature overlaps with numerical characteristic in the statistical nature of wind power plant predicated error；Pass through K-means clustering algorithms if overlapping
Clustered again；
Prediction module, is distributed for the wind power prediction result according to each moment and condition experience, obtains wind power general
Rate predicts the outcome.
9. wind power probabilistic forecasting device according to claim 8, it is characterised in that the feature calculation module bag
Include：
First acquisition unit, for obtaining the history power output of wind power plant within a predetermined period of time；
Second acquisition unit, the historical forecast power for obtaining wind power plant within a predetermined period of time；
First computing unit, for calculate the error in predetermined amount of time between history power output and historical forecast power to
Amount；
Second computing unit, for obtaining each component in error vector, constitutes wind power plant predicated error sample value, calculates wind-powered electricity generation
The statistical nature of field prediction error.
10. wind power probabilistic forecasting device as claimed in claim 8, it is characterised in that the condition computing module includes：
Undulate quantity acquiring unit, for the historical forecast power of wind power plant in the predetermined amount of time according to acquisition, is obtained in moment t
When, the fluctuations in wind speed amount that wind power plant NWP predicts the outcome；
Duration set acquiring unit is fluctuated, for obtaining the fluctuations in wind speed quantity set that the fluctuations in wind speed amount at the moment from t-s to t-1 is constituted
Close；
Condition complete or collected works' acquiring unit, is mutually combined the Descartes constituted collection by fluctuations in wind speed amount and the set for obtaining, made
For condition complete or collected works.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710055611.XA CN107067099B (en) | 2017-01-25 | 2017-01-25 | Wind power probability prediction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710055611.XA CN107067099B (en) | 2017-01-25 | 2017-01-25 | Wind power probability prediction method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107067099A true CN107067099A (en) | 2017-08-18 |
| CN107067099B CN107067099B (en) | 2020-06-19 |
Family
ID=59599172
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710055611.XA Active CN107067099B (en) | 2017-01-25 | 2017-01-25 | Wind power probability prediction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107067099B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108133279A (en) * | 2017-08-29 | 2018-06-08 | 甘肃省电力公司风电技术中心 | Wind power probability forecasting method, storage medium and equipment |
| CN108898251A (en) * | 2018-06-29 | 2018-11-27 | 上海电力学院 | Consider the marine wind electric field power forecasting method of meteorological similitude and power swing |
| CN109784563A (en) * | 2019-01-18 | 2019-05-21 | 南方电网科学研究院有限责任公司 | A kind of ultra-short term power forecasting method based on virtual anemometer tower technology |
| CN109840308A (en) * | 2017-11-29 | 2019-06-04 | 中国电力科学研究院有限公司 | A kind of region wind power probability forecast method and system |
| CN111310121A (en) * | 2019-11-14 | 2020-06-19 | 中国电力科学研究院有限公司 | New energy output probability prediction method and system |
| CN112232886A (en) * | 2020-10-30 | 2021-01-15 | 南方电网能源发展研究院有限责任公司 | Electricity price probability prediction method, system, computer equipment and storage medium |
| CN112270439A (en) * | 2020-10-28 | 2021-01-26 | 国能日新科技股份有限公司 | Ultra-short-term wind power prediction method and device, electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105022912A (en) * | 2015-05-28 | 2015-11-04 | 北京交通大学 | Rolling bearing fault prediction method based on wavelet principal component analysis |
| WO2015173803A2 (en) * | 2014-05-11 | 2015-11-19 | Ofek - Eshkolot Research And Development Ltd | A system and method for generating detection of hidden relatedness between proteins via a protein connectivity network |
| CN105389634A (en) * | 2015-12-01 | 2016-03-09 | 广东智造能源科技研究有限公司 | Combined short-term wind power prediction system and method |
-
2017
- 2017-01-25 CN CN201710055611.XA patent/CN107067099B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015173803A2 (en) * | 2014-05-11 | 2015-11-19 | Ofek - Eshkolot Research And Development Ltd | A system and method for generating detection of hidden relatedness between proteins via a protein connectivity network |
| CN105022912A (en) * | 2015-05-28 | 2015-11-04 | 北京交通大学 | Rolling bearing fault prediction method based on wavelet principal component analysis |
| CN105389634A (en) * | 2015-12-01 | 2016-03-09 | 广东智造能源科技研究有限公司 | Combined short-term wind power prediction system and method |
Non-Patent Citations (1)
| Title |
|---|
| 王松岩: "风速与风电功率的联合条件概率预测方法", 《中国电机工程学报》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108133279A (en) * | 2017-08-29 | 2018-06-08 | 甘肃省电力公司风电技术中心 | Wind power probability forecasting method, storage medium and equipment |
| CN108133279B (en) * | 2017-08-29 | 2020-12-22 | 甘肃省电力公司风电技术中心 | Wind power probability prediction method, storage medium and device |
| CN109840308A (en) * | 2017-11-29 | 2019-06-04 | 中国电力科学研究院有限公司 | A kind of region wind power probability forecast method and system |
| CN109840308B (en) * | 2017-11-29 | 2023-12-19 | 中国电力科学研究院有限公司 | Regional wind power probability forecasting method and system |
| CN108898251A (en) * | 2018-06-29 | 2018-11-27 | 上海电力学院 | Consider the marine wind electric field power forecasting method of meteorological similitude and power swing |
| CN109784563A (en) * | 2019-01-18 | 2019-05-21 | 南方电网科学研究院有限责任公司 | A kind of ultra-short term power forecasting method based on virtual anemometer tower technology |
| CN111310121A (en) * | 2019-11-14 | 2020-06-19 | 中国电力科学研究院有限公司 | New energy output probability prediction method and system |
| CN112270439A (en) * | 2020-10-28 | 2021-01-26 | 国能日新科技股份有限公司 | Ultra-short-term wind power prediction method and device, electronic equipment and storage medium |
| CN112270439B (en) * | 2020-10-28 | 2024-03-08 | 国能日新科技股份有限公司 | Ultra-short-term wind power prediction method and device, electronic equipment and storage medium |
| CN112232886A (en) * | 2020-10-30 | 2021-01-15 | 南方电网能源发展研究院有限责任公司 | Electricity price probability prediction method, system, computer equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107067099B (en) | 2020-06-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107067099A (en) | Wind power probability forecasting method and device | |
| Liu et al. | A systematic approach for dynamic security assessment and the corresponding preventive control scheme based on decision trees | |
| Tascikaraoglu et al. | A review of combined approaches for prediction of short-term wind speed and power | |
| Wang et al. | Decision tree based online stability assessment scheme for power systems with renewable generations | |
| US10379146B2 (en) | Detecting non-technical losses in electrical networks based on multi-layered statistical techniques from smart meter data | |
| US9196011B2 (en) | Smoothing power output from a wind farm | |
| JP6348862B2 (en) | System stabilization control device and power system control system | |
| Buwei et al. | A solar power prediction using support vector machines based on multi-source data fusion | |
| CN110472762A (en) | A kind of correction method and system of gridding precipitation | |
| CN103718129B (en) | For determining the method and apparatus of the power of being exported by photovoltaic apparatus | |
| CN111754024B (en) | Prediction method and system of power industry time series based on regression analysis | |
| CN108133279A (en) | Wind power probability forecasting method, storage medium and equipment | |
| CA2867253A1 (en) | System and method for analyzing oscillatory stability in electrical power transmission systems | |
| Thatte et al. | Robust optimization based economic dispatch for managing system ramp requirement | |
| CN109543083B (en) | Method for detecting abnormal data in real-time data of multi-element power grid | |
| CN104809532A (en) | Method for predicting generating capacity of photovoltaic system | |
| Ye et al. | A hybrid ARIMA-DENFIS method for wind speed forecasting | |
| CN108256693A (en) | A kind of photovoltaic power generation power prediction method, apparatus and system | |
| Saini et al. | Gated recurrent unit (gru) based short term forecasting for wind energy estimation | |
| Yu et al. | On statistical modeling and forecasting of energy usage in smart grid | |
| CN111160386B (en) | Data-driven power island detection method based on active reactive power disturbance | |
| CN116316617A (en) | Multi-station intelligent fusion new energy generation power region prediction method and system | |
| CN109816149A (en) | A kind of wind power plant is contributed scene generating method and device at random | |
| CN115907307A (en) | Power grid real-time data interaction-oriented online analysis method for carbon emission flow of power system | |
| Chen et al. | Machine learning forecasts of Scandinavian numerical weather prediction wind model residuals with control theory for wind energy |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |