CN113988360A - Wind power prediction method and device based on wind speed fluctuation characteristic typing - Google Patents

Wind power prediction method and device based on wind speed fluctuation characteristic typing Download PDF

Info

Publication number
CN113988360A
CN113988360A CN202111061677.2A CN202111061677A CN113988360A CN 113988360 A CN113988360 A CN 113988360A CN 202111061677 A CN202111061677 A CN 202111061677A CN 113988360 A CN113988360 A CN 113988360A
Authority
CN
China
Prior art keywords
fluctuation
wind speed
wind
power
fluctuation process
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202111061677.2A
Other languages
Chinese (zh)
Other versions
CN113988360B (en
Inventor
李强
孙蓉
吴盛军
贾勇勇
唐伟佳
吕振华
蒋明慧
卜京
刘伟伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing University of Science and Technology
State Grid Jiangsu Electric Power Co Ltd
Electric Power Research Institute of State Grid Jiangsu Electric Power Co Ltd
Original Assignee
Nanjing University of Science and Technology
State Grid Jiangsu Electric Power Co Ltd
Electric Power Research Institute of State Grid Jiangsu Electric Power Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nanjing University of Science and Technology, State Grid Jiangsu Electric Power Co Ltd, Electric Power Research Institute of State Grid Jiangsu Electric Power Co Ltd filed Critical Nanjing University of Science and Technology
Priority to CN202111061677.2A priority Critical patent/CN113988360B/en
Publication of CN113988360A publication Critical patent/CN113988360A/en
Application granted granted Critical
Publication of CN113988360B publication Critical patent/CN113988360B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • 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
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Landscapes

  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Engineering & Computer Science (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Development Economics (AREA)
  • Marketing (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Educational Administration (AREA)
  • Game Theory and Decision Science (AREA)
  • Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Water Supply & Treatment (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Wind Motors (AREA)

Abstract

The invention discloses a wind power prediction method and a wind power prediction device based on wind speed fluctuation characteristic typing. According to the method, the relevance between the wind speed fluctuation characteristics and the prediction model is considered for parting prediction, and the combination model is established to improve the universality of the model, so that the prediction precision of the wind power is improved.

Description

Wind power prediction method and device based on wind speed fluctuation characteristic typing
Technical Field
The invention relates to a wind power prediction method and device based on wind speed fluctuation characteristic typing, and belongs to the technical field of wind power prediction.
Background
The strong fluctuation, randomness and intermittence of wind energy determine that the wind power has strong fluctuation, after the wind power is merged into a power grid on a large scale, the fluctuation characteristic will bring huge challenges to safe and stable operation of the power grid, however, fluctuation change of the wind speed is not completely random, and still has certain regularity, so that the high-precision prediction of the wind speed and the wind power is very important.
At present, the main methods for predicting the domestic short-term wind power are mostly based on a time sequence method, an artificial intelligence algorithm and the like, models are established through linear or nonlinear relations between historical wind speed and power data, and then the wind power is predicted, however, the methods neglect the association between wind speed fluctuation characteristics and prediction models, and the prediction error of a single prediction model is large.
Disclosure of Invention
In order to solve the problems that the accuracy of a prediction model is low and the like due to the fact that the correlation between the fluctuation characteristic of the wind speed and the prediction model is ignored and single model prediction is adopted in the existing wind power prediction method, the invention provides a wind power prediction method and device based on wind speed fluctuation characteristic typing.
The technical scheme adopted by the invention is as follows:
the invention provides a wind power prediction method based on wind speed fluctuation characteristic typing, which comprises the following steps:
classifying the wind speed fluctuation characteristics based on the historical wind speed data of the weather forecast; classifying the wind power fluctuation characteristics based on historical wind power data of the wind power plant set; one of the fluctuation processes is: starting from a local minimum value smaller than a certain wind speed/wind power threshold value, sequentially passing through at least one wave crest, and returning to the local minimum value smaller than the wind speed/wind power threshold value to end;
dividing different types of weather processes based on matching between the wind speed fluctuation characteristic type and the wind power fluctuation characteristic type;
respectively establishing wind power prediction models for different types of weather processes, and training the established models based on historical data;
inputting the real-time wind speed data into a wind power prediction model corresponding to a weather process according to the wind speed fluctuation characteristic type;
and performing time sequence recombination on the wind power prediction results of different weather processes to obtain a wind power combination prediction value.
Further, in the above-mentioned case,
and carrying out normalization processing on the historical wind speed data and the historical wind power data.
Further, the classifying the wind speed fluctuation characteristics based on the historical wind speed data of the weather forecast includes:
Figure BDA0003256613210000021
wherein，
Figure BDA0003256613210000022
Is the maximum peak value in the single wind speed fluctuation process, lambda is 1,2, …, is the peak sequence,
Figure BDA0003256613210000023
is the peak value of the lambda-th wave peak, epsilonv,0For normalized wind farm cut-in wind speed, εv,1And εv,2Respectively, a threshold value for the determination of the type of the wind speed fluctuation process, Wv,0，Wv,1，Wv,2And Wv,3The method comprises a zero-output wind speed fluctuation process, a wind speed small fluctuation process, a wind speed medium fluctuation process and a wind speed large fluctuation process.
Further, the epsilonv,1∈[0.2，0.4]，εv,2∈[0.4，0.6]。
Further, classifying the wind power fluctuation characteristics based on historical wind power data of the wind power plant set includes:
Figure BDA0003256613210000024
wherein,
Figure BDA0003256613210000025
is the wave peak value in the single power fluctuation process, alpha is 1,2, …, alpha is the wave peak sequence,
Figure BDA0003256613210000026
is the peak value of the alpha peak, epsilonp,1And εp,2Respectively, a decision threshold, W, of different power fluctuation process typesp,1，Wp,2And Wp,3The method comprises a small power fluctuation process, a medium power fluctuation process and a large power fluctuation process.
Further, the epsilonv,1∈[0.2，0.4]，εv,2∈[0.4，0.6]。
Further, the classifying different types of weather processes based on the matching between the wind speed fluctuation feature type and the wind power fluctuation feature type includes:
the small fluctuation process of the wind speed is matched with the small fluctuation process of the power and divided into small fluctuation weather processes;
the fluctuation process in the wind speed is matched with the fluctuation process in the power and divided into medium fluctuation weather processes;
the large fluctuation process of the wind speed is matched with the large fluctuation process of the power, and the large fluctuation process is divided into large fluctuation weather processes.
Further, the respectively establishing wind power prediction models for different types of weather processes includes:
for a small fluctuation weather process, establishing a prediction model based on wind speed data of a small fluctuation process of wind speed and wind power data of a small fluctuation process of power by adopting an autoregressive moving average method;
and establishing a prediction model based on wind speed data in the fluctuation process of wind speed and wind power in the fluctuation process of power, and wind speed data in the large fluctuation process of wind speed and wind power data in the large fluctuation process of power by adopting a least square support vector machine method in the middle fluctuation weather process and the large fluctuation weather process.
The invention also provides a wind power prediction device based on wind speed fluctuation characteristic typing, which comprises the following components:
the classification module is used for classifying the wind speed fluctuation characteristics based on the historical wind speed data of the weather forecast; classifying the wind power fluctuation characteristics based on historical wind power data of the wind power plant set; one of the fluctuation processes is: starting from a local minimum value smaller than a certain wind speed/wind power threshold value, sequentially passing through at least one wave crest, and returning to the local minimum value smaller than the wind speed/wind power threshold value to end;
the matching module is used for dividing different types of weather processes based on matching between the wind speed fluctuation characteristic type and the wind power fluctuation characteristic type;
the model module is used for respectively establishing wind power prediction models for different types of weather processes and training the established models based on historical data;
and the number of the first and second groups,
and the recombination module is used for carrying out time sequence recombination on the wind power prediction results of different types of weather processes to obtain a wind power combination prediction value.
Further, the classification module is specifically configured to,
the wind speed fluctuation characteristics are classified as follows:
Figure BDA0003256613210000031
wherein,
Figure BDA0003256613210000032
is the maximum peak value in the single wind speed fluctuation process, lambda is 1,2, …, is the peak sequence,
Figure BDA0003256613210000033
is the peak value of the lambda-th wave peak, epsilonv,0For normalized wind farm cut-in wind speed, εv,1And εv,2Respectively, a threshold value for the determination of the type of the wind speed fluctuation process, Wv,0，Wv,1，Wv,2And Wv,3Respectively performing a zero-output wind speed fluctuation process, a wind speed small fluctuation process, a wind speed medium fluctuation process and a wind speed large fluctuation process;
the wind power fluctuation characteristics are classified as follows:
Figure BDA0003256613210000034
wherein,
Figure BDA0003256613210000035
is the wave peak value in the single power fluctuation process, alpha is 1,2, …, alpha is the wave peak sequence,
Figure BDA0003256613210000036
is the peak value of the alpha peak, epsilonp,1And εp,2Respectively, determination of different power fluctuation process typesOff threshold, Wp,1，Wp,2And Wp,3The method comprises a small power fluctuation process, a medium power fluctuation process and a large power fluctuation process.
Further, the matching module is specifically configured to,
matching the small fluctuation process of the wind speed with the small fluctuation process of the power, and dividing the small fluctuation process into small fluctuation weather processes;
matching the fluctuation process in the wind speed with the fluctuation process in the power, and dividing the fluctuation process into medium fluctuation weather processes;
the large fluctuation process of wind speed is matched with the large fluctuation process of power, and the large fluctuation process is divided into large fluctuation weather processes.
Further, the model module is specifically configured to,
for a small fluctuation weather process, establishing a prediction model based on wind speed data of a small fluctuation process of wind speed and wind power data of a small fluctuation process of power by adopting an autoregressive moving average method;
and establishing a prediction model based on wind speed data in the fluctuation process of wind speed and wind power in the fluctuation process of power, and wind speed data in the large fluctuation process of wind speed and wind power data in the large fluctuation process of power by adopting a least square support vector machine method in the middle fluctuation weather process and the large fluctuation weather process.
The invention has the beneficial effects that:
according to the method, wind power prediction models are respectively established for different types of weather processes based on wind speed fluctuation characteristic analysis, and the wind power prediction results of the different types of weather processes are subjected to time sequence recombination to obtain the wind power combination prediction value. According to the method, the relevance between the wind speed fluctuation characteristics and the prediction model is considered for parting prediction, and the combination model is established to improve the universality of the model, so that the prediction precision of the wind power is improved.
Drawings
FIG. 1 is a flow chart of the wind power prediction overall implementation of the invention based on wind speed fluctuation feature typing;
FIG. 2 is a flow chart of ARMA model building in the present invention;
fig. 3 is a flow chart of the LSSVM algorithm of the present invention.
Detailed Description
The invention is further described below. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present invention is not limited thereby.
The invention provides a wind power prediction method based on wind speed fluctuation characteristic typing. The method includes the steps that influences of weather fluctuation characteristics on wind power prediction accuracy are considered, fluctuation processes of wind speed and wind power are divided by analyzing fluctuation characteristics of the wind speed and the wind power, the weather fluctuation processes are classified based on relevance of wind speed and wind power fluctuation degrees, and on the basis, prediction models suitable for different weather fluctuation processes are built to reduce prediction errors. The specific implementation steps, as shown in fig. 1, include:
selecting wind speed which is one of main meteorological factors in meteorological characteristic factors of numerical weather forecast as a basic meteorological factor for dividing a weather fluctuation process, analyzing periodicity and regularity of historical wind speed fluctuation, and classifying wind speed fluctuation types based on identification of fluctuation characteristics of historical wind speed;
and researching the wind power fluctuation characteristics based on the historical operation data of the wind power plant, and classifying the fluctuation type of the wind power.
Specifically, firstly, the historical wind speed/wind power data is normalized, so that the subsequent fluctuation characteristic identification is facilitated;
secondly, defining a fluctuation process of wind speed/wind power as follows: starting from a local minimum value smaller than a certain wind speed/wind power threshold value, sequentially passing through a single or a plurality of wave crests, and then returning to the local minimum value smaller than the wind speed/wind power threshold value to end, taking a wind speed fluctuation process as an example, the mathematical expression form is as follows:
Figure BDA0003256613210000041
wherein,
Figure BDA0003256613210000051
and
Figure BDA0003256613210000052
respectively, the start and end values of a fluctuation process,
Figure BDA0003256613210000053
is composed of
Figure BDA0003256613210000054
And
Figure BDA0003256613210000055
the wind power plant cut-in wind speed epsilon is smaller than or equal to the normalized wind power plant cut-in wind speed epsilonv,0，
Figure BDA0003256613210000056
The wave peak value in a single wind speed fluctuation process is represented, lambda is a peak value sequence, n (-) is a statistical function of the number of peak values, and a complete wind speed fluctuation process at least comprises one fluctuation peak value.
The amplitude, the fluctuation duration and the number of wave peaks of the wind speed/wind power fluctuation process divided based on the above rules are large, so that the wind speed/wind power fluctuation process needs to be classified according to specific parameters, and the wind speed fluctuation process can be expressed as follows by dividing into examples:
Figure BDA0003256613210000057
wherein,
Figure BDA0003256613210000058
is the maximum peak value, epsilon, in the course of a single wind speed fluctuationv,1And εv,2Respectively, a threshold value for the determination of the type of the wind speed fluctuation process, Wv,0，Wv,1，Wv,2，Wv,3The method comprises a zero-output wind speed fluctuation process, a wind speed small fluctuation process, a wind speed medium fluctuation process and a wind speed large fluctuation process. Reference to related document,. epsilonv,1∈[0.2，0.4]，εv,2∈[0.4，0.6]Considering the specific conditions of the wind power plant, epsilon can be selectedv,1＝0.25，εv,2＝0.5。
The division principle of the wind power fluctuation process is as above, and can be expressed as follows:
Figure BDA0003256613210000059
wherein,
Figure BDA00032566132100000510
is the wave peak value in the single power fluctuation process; alpha is a peak value sequence, similar to the wind speed peak value sequence, a complete power fluctuation process at least comprises one fluctuation peak value, and the actual power fluctuation process only comprises one peak value; epsilonp,1And εp,2Respectively, a decision threshold, W, of different power fluctuation process typesp,1，Wp,2，Wp,3The method comprises a small power fluctuation process, a medium power fluctuation process and a large power fluctuation process. Reference to related document,. epsilonp,1∈[0.2，0.4]，εp,2∈[0.4，0.6]Considering the specific conditions of the wind power plant, epsilon can be selectedp,1＝0.25，εp,2＝0.5。
Step two, carrying out statistics on the correlation between the wind speed fluctuation characteristics and the wind power fluctuation characteristics of the wind power plant in the same period, and dividing different types of weather processes according to the fluctuation rule matching;
specifically, wind power fluctuation types in different wind speed fluctuation processes in a selected wind power plant are statistically analyzed within a period of time, wherein a zero output power section corresponding to a zero output wind speed fluctuation process is not considered and is obtained through analysis according to a statistical result; the wind speed small fluctuation process of the wind power plant is mainly matched with the power small fluctuation process, the fluctuation process in the wind speed is mainly matched with the fluctuation process in the power, the large power fluctuation process is often matched, and the wind speed large fluctuation process is mainly matched with the power large fluctuation process.
Step three, establishing a wind power prediction method suitable for different weather processes according to the division result;
specifically, fluctuation characteristics under different weather fluctuation types are analyzed respectively, and a prediction model corresponding to the fluctuation characteristics is established. For the small weather fluctuation type corresponding to the small wind speed fluctuation type, an Autoregressive moving average (ARMA) method is selected for prediction, and the method is mainly characterized in that the Autoregressive moving average method is simple in model, high in calculation efficiency and suitable for prediction of wind power under the weather types with small wind speed and small fluctuation amplitude; for the weather large fluctuation type corresponding to the fluctuation type in the wind speed and the weather large fluctuation type corresponding to the wind speed large fluctuation type, a Least Square Support Vector Machine (LSSVM) is selected for prediction, and the method is mainly characterized in that the training speed of the LSSVM is high, the generalization capability is strong, and the method is suitable for the weather fluctuation type with large fluctuation amplitude and large sample size.
An autoregressive moving average model ARMA (n, m) is established for the output power data of the wind turbine generator set as follows:
Figure BDA0003256613210000061
the model comprises n autoregressive terms and m moving average terms, wherein n and m are the autoregressive order and the moving average order of the model respectively;
Figure BDA0003256613210000062
and theta are both undetermined coefficients other than zero,
Figure BDA0003256613210000063
as autoregressive parameter, θj(j ═ 1,2, …, n) is a moving average parameter; { alpha ]tIs a sequence of individual error terms, the mean of which is zero; x is the number oftIs a smooth time sequence inIn this model, xtWind speed at time t, xt-iIs the wind speed at time t-i.
The specific prediction process for ARMA is shown in fig. 2, as follows:
(1) judging whether the input time sequence is stable or not, and if not, carrying out differential processing to enable the input time sequence to tend to zero mean value stabilization;
(2) establishing an ARMA model, gradually increasing the order of the model, fitting the ARMA (n, n-1) model, estimating the model parameters by adopting a nonlinear least square method, selecting the model corresponding to the minimum variance of the residual sequence as a primary selection model, and determining the order;
(3) and (3) performing model adaptability test, judging whether the model passes parameter significance test and residual error test, if not, re-determining the difference order and repeating the step (2), and if so, finishing model training and starting prediction.
The LSSVM maps the samples to a high-dimensional feature space through a nonlinear mapping function, and records the nonlinear relation between input and output of training samples, so that the prediction precision can be effectively improved when the training data and the prediction data are of the same type. The nonlinear function of LSSVM is:
Figure BDA0003256613210000064
wherein y (x) is the output vector of the model, i.e. the wind power value, xi(i is 1,2, …, t) is a correlation vector of the prediction result, namely a wind speed value in the model, and t is the total number of time points of the training set; a isiIs Lagrange multiplier, b is offset, K (x, x)i) And as a kernel function, all three need to solve an optimal value in the training process.
The specific prediction steps of LSSVM are shown in fig. 3, as follows:
firstly, standardizing historical wind speed and wind power data, and normalizing all values with different dimension indexes to [ -1, 1 ];
secondly, respectively selecting proper LSSVM kernel functions and parameters for training according to the fluctuation rule of the wind speed;
thirdly, selecting a proper kernel function and a corresponding parameter according to the error size obtained by the training data and a cross validation method;
step four, training again by using the selected kernel function and parameters, checking the prediction result, completing model establishment if the prediction result passes the checking, and repeating the step three if the prediction result does not pass the checking;
and fifthly, inputting the real-time wind speed data into the trained model for prediction to obtain a final prediction result.
Performing time sequence recombination on the prediction results of different models to obtain a wind power combination prediction value;
based on the classification of the wind speed time sequence, different prediction models are respectively selected according to the fluctuation rule of the wind speed, and finally, the separated model prediction values are combined in time sequence to obtain the final output result.
Another embodiment of the present invention provides a wind power prediction apparatus based on wind speed fluctuation feature classification, including:
the classification module is used for classifying the wind speed fluctuation characteristics based on the historical wind speed data of the weather forecast; classifying the wind power fluctuation characteristics based on historical wind power data of the wind power plant set; one of the fluctuation processes is: starting from a local minimum value smaller than a certain wind speed/wind power threshold value, sequentially passing through at least one wave crest, and returning to the local minimum value smaller than the wind speed/wind power threshold value to end;
the matching module is used for dividing different types of weather processes based on matching between the wind speed fluctuation characteristic type and the wind power fluctuation characteristic type;
the model module is used for respectively establishing wind power prediction models for different types of weather processes and training the established models based on historical data;
and the number of the first and second groups,
and the recombination module is used for carrying out time sequence recombination on the wind power prediction results of different types of weather processes to obtain a wind power combination prediction value.
In the embodiment of the present invention, the classification module is specifically configured to,
the wind speed fluctuation characteristics are classified as follows:
Figure BDA0003256613210000071
wherein,
Figure BDA0003256613210000072
is the maximum peak value in the single wind speed fluctuation process, lambda is 1,2, …, is the peak sequence,
Figure BDA0003256613210000073
is the peak value of the lambda-th wave peak, epsilonv,0For normalized wind farm cut-in wind speed, εv,1And εv,2Respectively, a threshold value for the determination of the type of the wind speed fluctuation process, Wv,0，Wv,1，Wv,2And Wv,3Respectively performing a zero-output wind speed fluctuation process, a wind speed small fluctuation process, a wind speed medium fluctuation process and a wind speed large fluctuation process;
the wind power fluctuation characteristics are classified as follows:
Figure BDA0003256613210000074
wherein,
Figure BDA0003256613210000081
is the wave peak value in the single power fluctuation process, alpha is 1,2, …, alpha is the wave peak sequence,
Figure BDA0003256613210000082
is the peak value of the alpha peak, epsilonp,1And εp,2Respectively, a decision threshold, W, of different power fluctuation process typesp,1，Wp,2And Wp,3The method comprises a small power fluctuation process, a medium power fluctuation process and a large power fluctuation process.
In the embodiment of the present invention, the matching module is specifically used for,
matching the small fluctuation process of the wind speed with the small fluctuation process of the power, and dividing the small fluctuation process into small fluctuation weather processes;
matching the fluctuation process in the wind speed with the fluctuation process in the power, and dividing the fluctuation process into medium fluctuation weather processes;
the large fluctuation process of wind speed is matched with the large fluctuation process of power, and the large fluctuation process is divided into large fluctuation weather processes.
In the embodiment of the present invention, the model module is specifically used for,
for a small fluctuation weather process, establishing a prediction model based on wind speed data of a small fluctuation process of wind speed and wind power data of a small fluctuation process of power by adopting an autoregressive moving average method;
and establishing a prediction model based on wind speed data in the fluctuation process of wind speed and wind power in the fluctuation process of power, and wind speed data in the large fluctuation process of wind speed and wind power data in the large fluctuation process of power by adopting a least square support vector machine method in the middle fluctuation weather process and the large fluctuation weather process.
It is to be noted that the apparatus embodiment corresponds to the method embodiment, and the implementation manners of the method embodiment are all applicable to the apparatus embodiment and can achieve the same or similar technical effects, so that the details are not described herein.
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.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.

Claims (12)

1. A wind power prediction method based on wind speed fluctuation characteristic typing is characterized by comprising the following steps:
classifying the wind speed fluctuation characteristics based on the historical wind speed data of the weather forecast; classifying the wind power fluctuation characteristics based on historical wind power data of the wind power plant set; one of the fluctuation processes is: starting from a local minimum value smaller than a certain wind speed/wind power threshold value, sequentially passing through at least one wave crest, and returning to the local minimum value smaller than the wind speed/wind power threshold value to end;
dividing different types of weather processes based on matching between the wind speed fluctuation characteristic type and the wind power fluctuation characteristic type;
respectively establishing wind power prediction models for different types of weather processes, and training the established models based on historical data;
inputting the real-time wind speed data into a wind power prediction model corresponding to a weather process according to the wind speed fluctuation characteristic type;
and performing time sequence recombination on the wind power prediction results of different weather processes to obtain a wind power combination prediction value.
2. The wind power prediction method based on wind speed fluctuation feature typing according to claim 1, further comprising:
and carrying out normalization processing on the historical wind speed data and the historical wind power data.
3. The wind power prediction method based on wind speed fluctuation feature typing of claim 2, wherein the classifying the wind speed fluctuation features based on the weather forecast historical wind speed data comprises:
Figure RE-FDA0003389622410000011
wherein,
Figure RE-FDA0003389622410000012
is the maximum peak value in the single wind speed fluctuation process, lambda is 1,2, …, is the peak sequence,
Figure RE-FDA0003389622410000013
is the peak value of the lambda-th wave peak, epsilonv,0For normalized wind farm cut-in wind speed, εv,1And εv,2Respectively, a threshold value for the determination of the type of the wind speed fluctuation process, Wv,0，Wv,1，Wv,2And Wv,3The method comprises a zero-output wind speed fluctuation process, a wind speed small fluctuation process, a wind speed medium fluctuation process and a wind speed large fluctuation process.
4. The wind power prediction method based on wind speed fluctuation feature typing according to claim 3, wherein epsilonv,1∈[0.2，0.4]，εv,2∈[0.4，0.6]。
5. The wind power prediction method based on wind speed fluctuation feature typing according to claim 1, wherein the classifying the wind power fluctuation features based on historical wind power data of wind power plants comprises:
Figure RE-FDA0003389622410000014
wherein,
Figure RE-FDA0003389622410000015
is the wave peak value in the single power fluctuation process, alpha is 1,2, …, alpha is the wave peak sequence,
Figure RE-FDA0003389622410000016
is the peak value of the alpha peak, epsilonp,1And εp,2Respectively, a decision threshold, W, of different power fluctuation process typesp,1，Wp,2And Wp,3The method comprises a small power fluctuation process, a medium power fluctuation process and a large power fluctuation process.
6. Wind based on wind speed fluctuation feature typing according to claim 5Electric power prediction method, characterized in thatv,1∈[0.2，0.4]，εv,2∈[0.4，0.6]。
7. The wind power prediction method based on wind speed fluctuation feature classification as claimed in claim 1, wherein the classifying different types of weather processes based on the matching between the wind speed fluctuation feature type and the wind power fluctuation feature type comprises:
the small fluctuation process of the wind speed is matched with the small fluctuation process of the power and divided into small fluctuation weather processes;
the fluctuation process in the wind speed is matched with the fluctuation process in the power and divided into medium fluctuation weather processes;
the large fluctuation process of the wind speed is matched with the large fluctuation process of the power, and the large fluctuation process is divided into large fluctuation weather processes.
8. The wind power prediction method based on wind speed fluctuation feature typing according to claim 7, wherein the respectively establishing wind power prediction models for different types of weather processes comprises:
for a small fluctuation weather process, establishing a prediction model based on wind speed data of a small fluctuation process of wind speed and wind power data of a small fluctuation process of power by adopting an autoregressive moving average method;
and establishing a prediction model based on wind speed data in the fluctuation process of wind speed and wind power in the fluctuation process of power, and wind speed data in the large fluctuation process of wind speed and wind power data in the large fluctuation process of power by adopting a least square support vector machine method in the middle fluctuation weather process and the large fluctuation weather process.
9. A wind power prediction device based on wind speed fluctuation feature typing is characterized by comprising:
the classification module is used for classifying the wind speed fluctuation characteristics based on the historical wind speed data of the weather forecast; classifying the wind power fluctuation characteristics based on historical wind power data of the wind power plant set; one of the fluctuation processes is: starting from a local minimum value smaller than a certain wind speed/wind power threshold value, sequentially passing through at least one wave crest, and returning to the local minimum value smaller than the wind speed/wind power threshold value to end;
the matching module is used for dividing different types of weather processes based on matching between the wind speed fluctuation characteristic type and the wind power fluctuation characteristic type;
the model module is used for respectively establishing wind power prediction models for different types of weather processes and training the established models based on historical data;
and the number of the first and second groups,
and the recombination module is used for carrying out time sequence recombination on the wind power prediction results of different types of weather processes to obtain a wind power combination prediction value.
10. The wind power prediction device based on wind speed fluctuation feature typing according to claim 9, wherein the classification module is specifically configured to,
the wind speed fluctuation characteristics are classified as follows:
Figure RE-FDA0003389622410000021
wherein,
Figure RE-FDA0003389622410000031
is the maximum peak value in the single wind speed fluctuation process, lambda is 1,2, …, is the peak sequence,
Figure RE-FDA0003389622410000032
is the peak value of the lambda-th wave peak, epsilonv,0For normalized wind farm cut-in wind speed, εv,1And εv,2Respectively, a threshold value for the determination of the type of the wind speed fluctuation process, Wv,0，Wv,1，Wv,2And Wv,3Respectively performing a zero-output wind speed fluctuation process, a wind speed small fluctuation process, a wind speed medium fluctuation process and a wind speed large fluctuation process;
the wind power fluctuation characteristics are classified as follows:
Figure RE-FDA0003389622410000033
wherein,
Figure RE-FDA0003389622410000034
is the wave peak value in the single power fluctuation process, alpha is 1,2, …, alpha is the wave peak sequence,
Figure RE-FDA0003389622410000035
is the peak value of the alpha peak, epsilonp,1And εp,2Respectively, a decision threshold, W, of different power fluctuation process typesp,1，Wp,2And Wp,3The method comprises a small power fluctuation process, a medium power fluctuation process and a large power fluctuation process.
11. The wind power prediction device based on wind speed fluctuation feature typing according to claim 10, wherein the matching module is specifically configured to,
matching the small fluctuation process of the wind speed with the small fluctuation process of the power, and dividing the small fluctuation process into small fluctuation weather processes;
matching the fluctuation process in the wind speed with the fluctuation process in the power, and dividing the fluctuation process into medium fluctuation weather processes;
the large fluctuation process of wind speed is matched with the large fluctuation process of power, and the large fluctuation process is divided into large fluctuation weather processes.
12. The wind power prediction device based on wind speed fluctuation feature typing according to claim 11, wherein the model module is specifically configured to,
for a small fluctuation weather process, establishing a prediction model based on wind speed data of a small fluctuation process of wind speed and wind power data of a small fluctuation process of power by adopting an autoregressive moving average method;
and establishing a prediction model based on wind speed data in the fluctuation process of wind speed and wind power in the fluctuation process of power, and wind speed data in the large fluctuation process of wind speed and wind power data in the large fluctuation process of power by adopting a least square support vector machine method in the middle fluctuation weather process and the large fluctuation weather process.
CN202111061677.2A 2021-09-10 2021-09-10 Wind power prediction method and device based on wind speed fluctuation characteristic typing Active CN113988360B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111061677.2A CN113988360B (en) 2021-09-10 2021-09-10 Wind power prediction method and device based on wind speed fluctuation characteristic typing

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111061677.2A CN113988360B (en) 2021-09-10 2021-09-10 Wind power prediction method and device based on wind speed fluctuation characteristic typing

Publications (2)

Publication Number Publication Date
CN113988360A true CN113988360A (en) 2022-01-28
CN113988360B CN113988360B (en) 2024-07-12

Family

ID=79735606

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111061677.2A Active CN113988360B (en) 2021-09-10 2021-09-10 Wind power prediction method and device based on wind speed fluctuation characteristic typing

Country Status (1)

Country Link
CN (1) CN113988360B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102880810A (en) * 2012-10-25 2013-01-16 山东电力集团公司电力科学研究院 Wind power prediction method based on time sequence and neural network method
CN105184391A (en) * 2015-08-19 2015-12-23 国网山东省电力公司电力科学研究院 Method for predicting wind speed and power of wind farm based on wavelet decomposition and support vector machine
CN111008725A (en) * 2019-10-23 2020-04-14 中国农业大学 Meteorological factor fluctuation feature extraction method for short-term wind power prediction
CN112200346A (en) * 2020-09-07 2021-01-08 中国农业大学 Short-term wind power prediction method for weather fluctuation process division and matching

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102880810A (en) * 2012-10-25 2013-01-16 山东电力集团公司电力科学研究院 Wind power prediction method based on time sequence and neural network method
CN105184391A (en) * 2015-08-19 2015-12-23 国网山东省电力公司电力科学研究院 Method for predicting wind speed and power of wind farm based on wavelet decomposition and support vector machine
CN111008725A (en) * 2019-10-23 2020-04-14 中国农业大学 Meteorological factor fluctuation feature extraction method for short-term wind power prediction
CN112200346A (en) * 2020-09-07 2021-01-08 中国农业大学 Short-term wind power prediction method for weather fluctuation process division and matching

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
党睿;张俊芳;: "基于自回归滑动平均模型的风电功率预测", 安徽工业大学学报(自然科学版), no. 03, 15 July 2015 (2015-07-15) *

Also Published As

Publication number Publication date
CN113988360B (en) 2024-07-12

Similar Documents

Publication Publication Date Title
CN111860982A (en) Wind power plant short-term wind power prediction method based on VMD-FCM-GRU
CN109886464B (en) Low-information-loss short-term wind speed prediction method based on optimized singular value decomposition generated feature set
CN111091236B (en) Multi-classification deep learning short-term wind power prediction method classified according to pitch angles
CN110717610A (en) Wind power prediction method based on data mining
CN113344288B (en) Cascade hydropower station group water level prediction method and device and computer readable storage medium
CN112186761B (en) Wind power scene generation method and system based on probability distribution
Zhang et al. A hybrid forecasting system with complexity identification and improved optimization for short-term wind speed prediction
CN103413038A (en) Vector quantization based long-term intuitionistic fuzzy time series prediction method
CN109902340A (en) It is a kind of meter and complicated meteorology coupled characteristic multi-source-lotus combine scene generating method
CN114266416A (en) Photovoltaic power generation power short-term prediction method and device based on similar days and storage medium
CN112651576A (en) Long-term wind power prediction method and device
CN114021483A (en) Ultra-short-term wind power prediction method based on time domain characteristics and XGboost
CN109921426A (en) Wind-electricity integration system probability load flow calculation method based on CV-KDE
CN116596129A (en) Electric vehicle charging station short-term load prediction model construction method
CN117828413A (en) Transformer oil temperature prediction method and system based on LSTM neural network
CN116632820A (en) Wind power prediction method based on sequence extension VMD and wind speed fluctuation characteristics
CN116245019A (en) Load prediction method, system, device and storage medium based on Bagging sampling and improved random forest algorithm
CN115660232A (en) Ultra-short-term prediction method, device and system for wind power
CN114186412A (en) Hydropower station water turbine top cover long sequence water level prediction system and method based on self-attention mechanism
Sun et al. A compound structure for wind speed forecasting using MKLSSVM with feature selection and parameter optimization
CN104834816A (en) Short-term wind speed prediction method
CN117374917A (en) Multiple regression prediction method and device for ultra-short-term power load
CN112132344A (en) Short-term wind power prediction method based on similar day and FRS-SVM
CN113988360B (en) Wind power prediction method and device based on wind speed fluctuation characteristic typing
CN113705920B (en) Method for generating water data sample set for thermal power plant and terminal equipment

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