CN107067099B - Wind power probability prediction method and device - Google Patents
Wind power probability prediction method and device Download PDFInfo
- Publication number
- CN107067099B CN107067099B CN201710055611.XA CN201710055611A CN107067099B CN 107067099 B CN107067099 B CN 107067099B CN 201710055611 A CN201710055611 A CN 201710055611A CN 107067099 B CN107067099 B CN 107067099B
- Authority
- CN
- China
- Prior art keywords
- wind power
- prediction
- wind
- error
- condition
- 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.)
- Active
Links
Images
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 invention relates to a wind power probability prediction method and a device, wherein the method comprises the following steps: acquiring statistical characteristics of prediction errors of the wind power plant according to the historical output power and the historical prediction power; obtaining a condition complete set of wind power probability prediction according to the wind speed fluctuation amount of the NWP prediction result of the historical predicted power and the wind power plant; dividing the condition complete set into a plurality of condition subsets through a K-means clustering algorithm; forming conditional experience distribution for the error set under each condition subset, and checking whether the digital characteristics of the conditional experience distribution coincide with the digital characteristics in the statistical characteristics of the wind power plant prediction errors; if the two groups are overlapped, clustering again through a K-means clustering algorithm; and obtaining a wind power probability prediction result according to the wind power prediction result and the condition empirical distribution at each moment. The invention also relates to a prediction device. The wind power probability prediction method provided by the invention can differentially provide error distribution functions and has higher prediction accuracy.
Description
Technical Field
The invention relates to a wind power probability prediction method, in particular to a wind power probability prediction method and a wind power probability prediction device.
Background
The randomness, the volatility and the uncertainty of wind resources and the uncontrollable property of wind power output bring great troubles to the safe and stable operation of a power system. Wind power prediction technology has become a necessary technology.
However, in actual operation, prediction errors in the wind power prediction result cannot be avoided, and conventional wind power prediction generally adopts wind power point prediction, so that wind power uncertainty information cannot be provided. Unpredictable prediction errors exist in traditional wind power prediction, so that the prediction is inaccurate, the safety and stability analysis of a power system is influenced, and the effectiveness of an operation decision result is influenced.
Disclosure of Invention
In summary, it is necessary to provide a method and an apparatus capable of accurately predicting a wind power probability.
A wind power probability prediction method comprises the following steps:
acquiring statistical characteristics of prediction errors of the wind power plant according to the historical output power and the historical prediction power;
obtaining a condition complete set of wind power probability prediction according to the wind speed fluctuation amount of the NWP prediction result of the historical predicted power and the wind power plant;
dividing the condition complete set into a plurality of condition subsets through a K-means clustering algorithm;
forming conditional experience distribution for the error set under each condition subset, and checking whether the digital characteristics of the conditional experience distribution coincide with the digital characteristics in the statistical characteristics of the wind power plant prediction errors; if the two groups are overlapped, clustering again through a K-means clustering algorithm; and
and obtaining a wind power probability prediction result according to the wind power prediction result and the condition empirical distribution at each moment.
In one embodiment, the statistical characteristics of the wind farm prediction error include a mean e (x), a variance var (x), a skewness skew (x), and a kurtosis (x) of the wind farm prediction error;
in one embodiment, the step of obtaining the statistical characteristics of the prediction error of the wind farm according to the historical output power and the historical predicted power includes:
acquiring historical output power of a wind power plant in a preset time period;
acquiring historical predicted power of a wind power plant in a preset time period;
calculating an error vector in a predetermined time period; and
and obtaining each component in the error vector to form a wind power plant prediction error sample value, and calculating the statistical characteristics of the wind power plant prediction error.
In one embodiment, the step of obtaining a condition complete set of wind power probability prediction according to the wind speed fluctuation amount of the NWP prediction result of the historical predicted power and wind farm comprises:
obtaining the wind speed fluctuation amount of the NWP prediction result of the wind power plant at the moment t according to the obtained historical prediction power of the wind power plant within the preset time period;
acquiring a wind speed fluctuation amount set formed by wind speed fluctuation amounts from t-s to t-1; and
and acquiring a Cartesian set formed by mutually combining the historical predicted power and the set as a conditional complete set.
In one embodiment, the step of dividing the condition complete set into a plurality of condition subsets by using a K-means clustering algorithm includes:
and for the condition complete set, dividing the condition complete set into m clusters by using a K-means clustering algorithm, namely dividing the condition complete set into m mutually-disjoint condition subsets, wherein the value of m is 4-10, and the number of elements in each condition subset is evenly distributed.
In one embodiment, the step of forming a conditional empirical distribution for the error sets under each condition subset and checking whether the digital characteristics of the conditional empirical distribution coincide with the digital characteristics of the statistical characteristics of the prediction errors of the wind farm includes:
for each condition subset CiObtaining error sample set E under the conditioniAnd calculating the empirical distribution PDF of the wind power prediction error under different conditionsiAnd describing the condition C by using the empirical distribution of the wind power prediction erroriError random variable e ofi:
ei～PDFi；
And (4) checking the conditional empirical distribution of the m errors one by one to determine whether the conditional empirical distribution coincides with an error empirical distribution function of the digital characteristics in the statistical characteristics of the wind power plant prediction errors.
In one embodiment, the step of obtaining the wind power probability prediction result according to the wind power prediction result and the conditional empirical distribution at each time includes:
obtaining the wind power prediction result of the moment t + k at the moment t
And the predicted wind speed fluctuation amount fluc at time t + kt+k；
Let real number pair
In conditional subset CjSelecting an empirical distribution of error PDF for the subset of conditionsjAnd the error is randomly varied by ejAs a random variable of error at time t + k;
wind power prediction result combined with t + k
Obtaining a final probability prediction result:
wherein the content of the first and second substances,
the wind power probability prediction result is a random variable and is used for describing uncertain information of wind power.
A wind power probability prediction device, the device comprising:
the characteristic calculation module is used for acquiring the statistical characteristics of the prediction error of the wind power plant according to the historical output power and the historical prediction power;
the condition calculation module is used for obtaining a condition complete set of wind power probability prediction according to the wind speed fluctuation quantity of the NWP prediction result of the historical predicted power and the wind power plant;
the classification module is used for dividing the condition complete set into a plurality of condition subsets through a K-means clustering algorithm;
the clustering module is used for forming conditional experience distribution on the error set under each condition subset and checking whether the digital characteristics of the conditional experience distribution coincide with the digital characteristics in the statistical characteristics of the prediction errors of the wind power plant; if the two groups are overlapped, clustering again through a K-means clustering algorithm;
and the prediction module is used for obtaining a wind power probability prediction result according to the wind power prediction result and the condition empirical distribution at each moment.
The feature calculation module includes:
the first acquisition unit is used for acquiring historical output power of the wind power plant in a preset time period;
the second obtaining unit is used for obtaining historical predicted power of the wind power plant in a preset time period;
a first calculation unit for calculating an error vector between the historical output power and the historical predicted power within a predetermined period of time;
and the second calculating unit is used for acquiring each component in the error vector, forming a wind power plant prediction error sample value and calculating the statistical characteristics of the wind power plant prediction error.
The condition calculation module includes:
the fluctuation quantity obtaining unit is used for obtaining the wind speed fluctuation quantity of the NWP prediction result of the wind power plant at the time t according to the obtained historical predicted power of the wind power plant in the preset time period;
a fluctuation amount set acquisition unit for acquiring a wind speed fluctuation amount set composed of wind speed fluctuation amounts from t-s to t-1;
a conditional ensemble acquisition unit for acquiring a cartesian set composed of the historical predicted power and the set combined with each other as a conditional ensemble.
The wind power probability prediction technology provided by the invention is based on the conditional probability distribution and the K-means clustering algorithm, can differentially provide an error distribution function, and has higher prediction accuracy.
Drawings
Fig. 1 is a flow chart of wind power probability prediction according to an embodiment of the present invention.
Detailed Description
The technical scheme of the invention is further detailed in the following description and the accompanying drawings in combination with specific embodiments.
In a large wind power cluster, the power outputs of a plurality of adjacent wind power plants have strong correlation, so that the abnormal data point correction of the wind power plant can be carried out by combining the measured wind speed value of the wind power plant i and the measured output power values of the plurality of adjacent strong-correlation wind power plants, and the abnormal data point of the wind power plant i at the moment t is corrected
Corrected to corrected power estimate based on wind speed and spatial correlation
Two key problems need to be solved for realizing the abnormal data point correction of the wind power plant i: first, determine the measured wind speed as
And the reasonable range of the output power value is obtained. When the equivalent power curve of the wind power plant is known, the reasonable range of the output power value is the wind speed
And the value range is determined by the intersection point of the equivalent power curve and the upper and lower boundaries. Secondly, determining a current power value of the wind farm i can be determined by performing conditional recursive sampling based on conditional probability distribution of a strongly correlated wind farm (for example, a correlation coefficient is greater than 0.5). The reason for only considering the strongly correlated wind farm here is that the conditional probability distribution dimension can be reduced, the occurrence of dimension disaster is avoided, and for wind farm data correction, considering the strongly correlated wind farm can satisfy the correction requirement.
Referring to fig. 1, a wind power probability prediction method based on a clustering algorithm and error analysis according to an embodiment of the present invention includes the following steps:
and step S10, obtaining the statistical characteristics of the wind power plant prediction error.
Specifically, in one embodiment, the obtaining of the statistical characteristics of the prediction error of the wind farm includes:
step S11, obtaining the historical output power of the wind power plant in a preset time period, and the historical output power is as follows:
wherein t is the current moment, and s is the historical data duration traced back forward.
Step S12, obtaining the historical predicted power of the wind power plant in a preset time period, and the historical predicted power is as follows:
step S13, calculating an error vector between the historical output power and the historical predicted power in a predetermined time period as:
and step S14, obtaining each component in the error vector, forming a wind power plant prediction error sample value, and calculating the statistical characteristics of the wind power plant prediction error.
Specifically, the statistical characteristics of the wind farm prediction error may include a mean e (x), a variance var (x), a skewness skew (x), and a kurtosis (x) of the wind farm prediction error. Can make X be
Calculating the statistical characteristics of the wind power plant prediction error by using a wind power plant prediction error sample set formed by each component: mean E (X), variance Var (X), skewness Skaw (X), and kurtosis (X).
Step S20, obtaining a condition corpus of wind power probability prediction according to the wind speed fluctuation amount of the NWP (numerical weather prediction) prediction result of the historical predicted power and wind farm.
Specifically, in one embodiment, the obtaining of the condition complete set of the wind power probability prediction and the scatter diagram includes the following steps:
step S21, according to the obtained historical predicted power of the wind power plant in the preset time period:
and obtaining the wind speed fluctuation quantity of the NWP prediction result of the wind power plant at the moment t, wherein the wind speed fluctuation quantity is as follows:
step S22, selecting the wind speed fluctuation quantity fluc from t-S to t-1t-s,…,fluct-1The formed wind speed fluctuation amount is Fluct,sWherein s is greater than 1 and s is less than t.
Step S23, obtaining the data of
And Fluct,sThe two are combined to form a Cartesian set which is used as a condition complete set C and can form a corresponding scatter diagram.
And step S30, dividing the condition complete set into a plurality of condition subsets through a K-means clustering algorithm.
The condition complete set C can be classified by a K-means clustering algorithm to form a condition complete set C1,C2,…,CmM condition subsets.
The K-means algorithm is a distance-based clustering algorithm, and the distance is used as an evaluation index of similarity, that is, the closer the distance between two objects is, the greater the similarity of the two objects is.
Specifically, for the condition complete set C, a K-means clustering algorithm is used to divide the condition complete set C into m clusters, namely, the condition complete set C is divided into m mutually disjoint condition subsets:
the value of m can be selected to be 4-10, and the number of elements in each condition subset is ensured to be relatively average according to different conditions, namely the number of elements in each condition subset is basically the same.
Step S40, for each condition subset CiForming conditional experience distribution by the lower error set, and checking whether the digital characteristics of the conditional experience distribution are highly overlapped with the digital characteristics in the statistical characteristics of the wind power plant prediction errors; if the heights are overlapped, the process returns to step S30, and clustering is performed again.
Specifically, in one embodiment, for each subset of conditions CiUnder the condition, an error sample set E can be obtainediAnd the empirical distribution PDF of the wind power prediction error under different conditions can be obtainediThis distribution is used to describe the condition CiError random variable e ofi:
ei～PDFi；
It is checked one by one whether the conditional empirical distribution of m errors is highly coincident with the empirical distribution function of errors in step S10. The term "height coincidence" means that the difference between the two digital features (mean, variance, skewness, kurtosis) is smaller than a preset threshold, the preset threshold can be selected as required, and the smaller the preset threshold is, the more the digital features coincide with each other.
If the difference values are highly overlapped, the condition difference information of the error distribution cannot be effectively stripped by the secondary clustering. And returning to the step S30, and changing the value of m to perform clustering again until none of the conditional empirical distributions is highly overlapped with the empirical distribution in the statistical characteristics of the wind farm prediction error.
And step S50, combining the wind power prediction results at all times to form a wind power probability prediction result.
Specifically, in one embodiment, the obtaining of the wind power probability prediction result includes:
step S51, obtaining the wind power prediction result at the moment t + k at the moment t
And the predicted wind speed fluctuation amount fluc at time t + kt+kWherein k is greater than 0;
step S52, set the real number pair
In conditional subset CjSelecting an error empirical distribution PDF for the subset of conditionsjRandom variation of error ejAs a random variable of error at time t + k;
step S53, combining wind power prediction results of t + k
Forming a final probabilistic prediction result:
wherein the content of the first and second substances,
the wind power probability prediction method is a result of wind power probability prediction, is a random variable and can describe uncertain information of wind power.
The wind power probability prediction method provided by the embodiment of the invention is based on the conditional probability distribution and the K-means clustering algorithm, can differentially provide error distribution functions, and has better prediction effect. According to the wind power probability prediction method, the prediction error of the wind power is modeled and analyzed, the possible fluctuation range and the probability distribution of the wind power can be comprehensively described, and therefore reliable and detailed information is provided for the wind power prediction result.
Based on the same inventive concept, the embodiment of the invention also provides a wind power probability prediction device, and as the principle of solving the problems of the devices is similar to a wind power probability prediction method, the implementation of the devices can be referred to the implementation of the method, and repeated parts are not described again.
The wind power probability prediction device may include:
the characteristic calculation module is used for acquiring the statistical characteristics of the prediction error of the wind power plant according to the historical output power and the historical prediction power;
the condition calculation module is used for obtaining a condition complete set of wind power probability prediction according to the wind speed fluctuation quantity of the NWP prediction result of the historical predicted power and the wind power plant;
the classification module is used for dividing the condition complete set into a plurality of condition subsets through a K-means clustering algorithm;
the clustering module is used for forming conditional experience distribution on the error set under each condition subset and checking whether the digital characteristics of the conditional experience distribution coincide with the digital characteristics in the statistical characteristics of the prediction errors of the wind power plant; if the two groups are overlapped, clustering again through a K-means clustering algorithm;
and the prediction module is used for obtaining a wind power probability prediction result according to the wind power prediction result and the condition empirical distribution at each moment.
In an implementation, the feature calculation module may include:
the first acquisition unit is used for acquiring historical output power of the wind power plant in a preset time period;
the second obtaining unit is used for obtaining historical predicted power of the wind power plant in a preset time period;
a first calculation unit for calculating an error vector between the historical output power and the historical predicted power within a predetermined period of time;
and the second calculating unit is used for acquiring each component in the error vector, forming a wind power plant prediction error sample value and calculating the statistical characteristics of the wind power plant prediction error.
In an implementation, the condition calculating module may include:
the fluctuation quantity obtaining unit is used for obtaining the wind speed fluctuation quantity of the NWP prediction result of the wind power plant at the time t according to the obtained historical predicted power of the wind power plant in the preset time period;
a fluctuation amount set acquisition unit, configured to acquire a wind speed fluctuation amount set formed by wind speed fluctuation amounts from t-s to t-1, where s is greater than 1 and s is less than t;
a conditional ensemble acquisition unit for acquiring a cartesian set composed of the historical predicted power and the set combined with each other as a conditional ensemble.
For convenience of description, each part of the above-described apparatus is separately described as being functionally divided into various modules or units. Of course, the functionality of the various modules or units may be implemented in the same one or more pieces of software or hardware when implementing the present application.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (9)
1. A wind power probability prediction method is characterized by comprising the following steps:
acquiring statistical characteristics of prediction errors of the wind power plant according to the historical output power and the historical prediction power;
according to the historical predicted power and the wind speed fluctuation amount of the numerical weather forecast NWP prediction result of the wind power plant, obtaining a condition complete set of wind power probability prediction;
dividing the condition complete set into a plurality of condition subsets through a K-means clustering algorithm;
for each condition subset CiObtaining error sample set E under the conditioniAnd calculating the empirical distribution PDF of the wind power prediction error under different conditionsiAnd describing the condition C by using the empirical distribution of the wind power prediction erroriError random variable e ofi：
ei～PDFi；
Checking whether the digital characteristics of the wind power plant prediction error coincide with the digital characteristics in the statistical characteristics of the wind power plant prediction error; if the two groups are overlapped, clustering again through a K-means clustering algorithm; and
obtaining the wind power prediction result of the moment t + k at the moment t
And the predicted wind speed fluctuation amount fluc at time t + kt+kWherein k is greater than 0;
let real number pair (
fluct+k) In conditional subset CjSelecting an empirical distribution of error PDF for the subset of conditionsjAnd the error is randomly varied by ejAs a random variable of error at time t + k;
wind power prediction result combined with t + k
Obtaining a final probability prediction result:
wherein the content of the first and second substances,
the wind power probability prediction result is a random variable and is used for describing uncertain information of wind power.
2. The wind power probability prediction method according to claim 1, characterized in that the statistical characteristics of the wind farm prediction error include a mean value e (x), a variance var (x), a skewness skew (x), and a kurtosis (x) of the wind farm prediction error.
3. The wind power probability prediction method according to claim 2, wherein the step of obtaining statistical characteristics of wind farm prediction errors according to historical output power and historical predicted power comprises:
acquiring historical output power of a wind power plant in a preset time period;
acquiring historical predicted power of a wind power plant in a preset time period;
calculating an error vector between historical output power and historical predicted power in a preset time period; and
and obtaining each component in the error vector to form a wind power plant prediction error sample value, and calculating the statistical characteristics of the wind power plant prediction error.
4. The wind power probability prediction method according to claim 1, wherein the step of obtaining the condition complete set of the wind power probability prediction according to the wind speed fluctuation amount of the numerical weather forecast NWP prediction result of the historical predicted power and the wind farm comprises:
obtaining the wind speed fluctuation amount of the NWP prediction result of the wind power plant at the moment t according to the obtained historical prediction power of the wind power plant within the preset time period;
acquiring a wind speed fluctuation amount set formed by wind speed fluctuation amounts from t-s to t-1, wherein s is smaller than t and is larger than 1; and
and acquiring a Cartesian set formed by mutually combining the historical predicted power and the wind speed fluctuation amount set as a condition complete set.
5. The wind power probability prediction method of claim 1, wherein the step of dividing the conditional corpus into a plurality of conditional subsets by a K-means clustering algorithm comprises:
and for the condition complete set, dividing the condition complete set into m clusters by using a K-means clustering algorithm, namely dividing the condition complete set into m mutually-disjoint condition subsets, wherein the value of m is 4-10, and the number of elements in each condition subset is evenly distributed.
6. The wind power probability prediction method according to claim 5, characterized in that the step of checking whether the digital features thereof coincide with the digital features in the statistical features of the wind farm prediction error comprises:
and (4) checking the conditional empirical distribution of the m errors one by one to determine whether the conditional empirical distribution coincides with an error empirical distribution function of the digital characteristics in the statistical characteristics of the wind power plant prediction errors.
7. A wind power probability prediction device, characterized in that the device comprises:
the characteristic calculation module is used for acquiring the statistical characteristics of the prediction error of the wind power plant according to the historical output power and the historical prediction power;
the condition calculation module is used for obtaining a condition complete set of wind power probability prediction according to the wind speed fluctuation quantity of the numerical weather forecast NWP prediction result of historical predicted power and a wind power plant;
the classification module is used for dividing the condition complete set into a plurality of condition subsets through a K-means clustering algorithm;
a clustering module for each condition subset CiObtaining error sample set E under the conditioniAnd calculating the empirical distribution PDF of the wind power prediction error under different conditionsiAnd describing the condition C by using the empirical distribution of the wind power prediction erroriError random variable e ofi：
ei～PDFi；
And checking whether the digital characteristics of the wind power plant prediction error coincide with the digital characteristics in the statistical characteristics of the wind power plant prediction error; if the two groups are overlapped, clustering again through a K-means clustering algorithm;
a prediction module for obtaining a wind power prediction result at the time t + k at the time t
And the predicted wind speed fluctuation amount fluc at time t + kt+kWherein k is greater than 0; let real number pair
fluct+k) In conditional subset CjSelecting an empirical distribution of error PDF for the subset of conditionsjAnd the error is randomly varied by ejAs an error random variable at the moment of t + k, and combining the wind power prediction result of t + k
Obtaining a final probability prediction result:
wherein the content of the first and second substances,
the wind power probability prediction result is a random variable and is used for describing uncertain information of wind power.
8. The wind power probability prediction device of claim 7, wherein the feature calculation module comprises:
the first acquisition unit is used for acquiring historical output power of the wind power plant in a preset time period;
the second obtaining unit is used for obtaining historical predicted power of the wind power plant in a preset time period;
a first calculation unit for calculating an error vector between the historical output power and the historical predicted power within a predetermined period of time;
and the second calculating unit is used for acquiring each component in the error vector, forming a wind power plant prediction error sample value and calculating the statistical characteristics of the wind power plant prediction error.
9. The wind power probability prediction device of claim 7, where the condition calculation module comprises:
the fluctuation quantity obtaining unit is used for obtaining the wind speed fluctuation quantity of the NWP prediction result of the wind power plant at the time t according to the obtained historical predicted power of the wind power plant in the preset time period;
a fluctuation amount set acquisition unit for acquiring a wind speed fluctuation amount set formed by wind speed fluctuation amounts from t-s to t-1, wherein s is less than t and s is greater than 1;
and a conditional ensemble acquisition unit configured to acquire a cartesian ensemble composed of the historical predicted power and the set of wind speed fluctuation amounts combined with each other as a conditional ensemble.
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 CN107067099A (en) | 2017-08-18 |
| CN107067099B true 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) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| 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 |
| CN108898251B (en) * | 2018-06-29 | 2021-10-08 | 上海电力学院 | Offshore wind farm power prediction method considering meteorological similarity and power fluctuation |
| CN109784563B (en) * | 2019-01-18 | 2023-05-23 | 南方电网科学研究院有限责任公司 | Ultra-short-term power prediction 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 |
| 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 |
| CN112232886B (en) * | 2020-10-30 | 2021-06-04 | 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 |
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 |
| 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 |
| WO2015173803A3 (en) * | 2014-05-11 | 2016-04-07 | Ofek - Eshkolot Research And Development Ltd | A system and method for generating detection of hidden relatedness between proteins via a protein connectivity network |
-
2017
- 2017-01-25 CN CN201710055611.XA patent/CN107067099B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015173803A3 (en) * | 2014-05-11 | 2016-04-07 | 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 |
|---|
| 风速与风电功率的联合条件概率预测方法;王松岩;《中国电机工程学报》;20110305;第31卷(第7期);第7-15页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107067099A (en) | 2017-08-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107067099B (en) | Wind power probability prediction method and device | |
| US9996444B2 (en) | Automated methods and systems for calculating hard thresholds | |
| CN108133279B (en) | Wind power probability prediction method, storage medium and device | |
| US9092725B2 (en) | Classifying attribute data intervals | |
| CN108228722B (en) | Method for detecting geographic space distribution uniformity of sampling points in crushing area | |
| CN105354198A (en) | Data processing method and apparatus | |
| CN111314173A (en) | Monitoring information abnormity positioning method and device, computer equipment and storage medium | |
| CN106452934B (en) | Method and device for analyzing network performance index change trend | |
| US20190087248A1 (en) | Anomaly detection and automated analysis using weighted directed graphs | |
| GB2603740A (en) | Alerting to model degradation based on distribution analysis using risk tolerance ratings | |
| CN105426998B (en) | Wind power interval prediction method based on multiple conditions | |
| CN110110339B (en) | Japanese hydrologic forecast error correction method and system | |
| Amarullah et al. | Planning decision support system using building mall AHP (Analytical Hierarchy Process) | |
| KR101343097B1 (en) | Fuzzy TOPSIS Approach method to Flood Vulnerability Assessment | |
| CN114328095A (en) | Task abnormity warning method and device | |
| US20120221272A1 (en) | Quality estimation apparatus, quality estimation method and non-transitory computer-readable medium storing program | |
| Omar et al. | Optimized feature selection based on a least-redundant and highest-relevant framework for a solar irradiance forecasting model | |
| EP3285206A1 (en) | Evaluation device, evaluation method and evaluation program for biometric data | |
| CN111262248B (en) | Random power flow analysis and calculation method and system | |
| CN111310121A (en) | New energy output probability prediction method and system | |
| CN112034139A (en) | Method and device for judging rock burst tendency grade and electronic equipment | |
| CN109840308B (en) | Regional wind power probability forecasting method and system | |
| CN106156470B (en) | Time series abnormity detection and labeling method and system | |
| CN109598644B (en) | Electricity stealing user identification method based on Gaussian distribution and terminal equipment | |
| Susetyoko et al. | Characteristics of Accuracy Function on Multiclass Classification Based on Best, Average, and Worst (BAW) Subset of Random Forest Model |
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 |