CN113824115B - Wind power frequency modulation energy prediction method and system and computer equipment - Google Patents

Wind power frequency modulation energy prediction method and system and computer equipment Download PDF

Info

Publication number
CN113824115B
CN113824115B CN202111398040.2A CN202111398040A CN113824115B CN 113824115 B CN113824115 B CN 113824115B CN 202111398040 A CN202111398040 A CN 202111398040A CN 113824115 B CN113824115 B CN 113824115B
Authority
CN
China
Prior art keywords
wind speed
historical
fan
data table
fans
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
Application number
CN202111398040.2A
Other languages
Chinese (zh)
Other versions
CN113824115A (en
Inventor
李少林
杨宁宁
秦世耀
代林旺
贺敬
张梅
李春彦
曲春辉
苗凤麟
李建立
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.)
China Electric Power Research Institute Co Ltd CEPRI
Original Assignee
China Electric Power Research Institute Co Ltd CEPRI
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 China Electric Power Research Institute Co Ltd CEPRI filed Critical China Electric Power Research Institute Co Ltd CEPRI
Priority to CN202111398040.2A priority Critical patent/CN113824115B/en
Publication of CN113824115A publication Critical patent/CN113824115A/en
Application granted granted Critical
Publication of CN113824115B publication Critical patent/CN113824115B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/004—Generation forecast, e.g. methods or systems for forecasting future energy generation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/24—Arrangements for preventing or reducing oscillations of power in networks
    • H02J3/241—The oscillation concerning frequency
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
    • H02J3/381—Dispersed generators
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2113/00—Details relating to the application field
    • G06F2113/04—Power grid distribution networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2113/00—Details relating to the application field
    • G06F2113/06—Wind turbines or wind farms
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
    • H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
    • H02J2300/20—The dispersed energy generation being of renewable origin
    • H02J2300/28—The renewable source being wind energy
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A30/00—Adapting or protecting infrastructure or their operation
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E10/00—Energy generation through renewable energy sources
    • Y02E10/70—Wind energy
    • Y02E10/76—Power conversion electric or electronic aspects

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Power Engineering (AREA)
  • Artificial Intelligence (AREA)
  • Software Systems (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Computational Linguistics (AREA)
  • Mathematical Physics (AREA)
  • General Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Computer Hardware Design (AREA)
  • Geometry (AREA)
  • Wind Motors (AREA)

Abstract

A wind power frequency modulation energy prediction method, a system and computer equipment comprise: respectively constructing a historical wind speed time sequence data table and a historical wind speed spatial data table based on the obtained historical wind speed data and the geographical position information of the fan within the determined historical duration, and respectively inputting the historical wind speed time sequence data table and the historical wind speed spatial data table into an LTC network link and a convolutional neural network link to obtain prediction information under consideration of mutual influence of the fans and prediction information under consideration of mutual influence of the fans; adding the prediction information considering the mutual influence of the fans and the prediction information considering the mutual influence of the fans, and then sequentially passing through two layers of full-connection and correction linear units to obtain a final prediction anemometer; and calculating the frequency modulation energy of the wind power plant based on the wind speed of each fan in the final predicted wind speed table at each moment. The method adopts the LTC network link and the convolutional neural network link to predict the wind speed of the fan, and achieves the effects of low calculation cost and high precision.

Description

Wind power frequency modulation energy prediction method and system and computer equipment
Technical Field
The invention relates to the technical field of new energy power generation, in particular to a wind power frequency modulation energy prediction method, a wind power frequency modulation energy prediction system and computer equipment.
Background
The wind speed/wind power frequency modulation energy prediction method can be roughly divided into a physical method, a statistical method and an intelligent method. The physical method has complex model and large calculation amount and is suitable for long-term prediction and wind field site selection; the statistical method has simple model and less data requirements, and is suitable for short-term prediction of the gray model; the intelligent method is suitable for ultra-short term prediction. Deep learning in an intelligent method is the most applied method at present, but a network model is large and the operation cost is high, so that a novel LTC recurrent neural network with small size, low test and calculation cost and high precision is bound to be the direction of time series data prediction; in addition, in recent years, the installed scale of wind power generation is rapidly increased, the penetration rate of a power grid of wind power generation is higher and higher, and the prediction of the frequency modulation energy of the wind power generation is more important.
The invention discloses a short-term wind speed prediction method of a wind power plant based on a hybrid neural network, aiming at a patent with the application number of CN201010557446.6 and named as 'short-term wind speed prediction method and system of the wind power plant based on the hybrid neural network', wherein the method comprises the steps of S1, determining input variables and output variables of a prediction model of the hybrid neural network according to preset prediction time intervals; and S2, carrying out wind speed prediction according to the hybrid neural network prediction model to obtain a corresponding wind speed prediction value. The invention also relates to a wind power plant short-term wind speed prediction system based on the hybrid neural network, which comprises a variable determination module: the device comprises a prediction model, a prediction model and a control module, wherein the prediction model is used for determining input variables and output variables of the hybrid neural network prediction model according to a preset prediction time interval; and a prediction module: and the wind speed prediction module is used for predicting the wind speed according to the hybrid neural network prediction model to obtain a corresponding wind speed prediction value. The wind power plant short-term wind speed prediction method and system based on the hybrid neural network have the advantages of high calculation speed and high reliability, solve the technical problem of completely depending on a physical prediction model, and overcome the defect of large prediction error fluctuation of a single model. The application is a method for mixing physical prediction and neural network prediction, and has large calculation amount and low precision.
Disclosure of Invention
In order to solve the problems of large network model and high operation cost in the prior art, the invention provides a wind power frequency modulation energy prediction method, which comprises the following steps:
respectively constructing a historical wind speed time sequence data table and a historical wind speed spatial data table based on the obtained historical wind speed data and the geographical position information of the fan within the determined historical duration, respectively taking the historical wind speed time sequence data table and the historical wind speed spatial data table as the input of an LTC network link and a convolutional neural network link, and respectively outputting prediction information under consideration of mutual influence of the fans and prediction information under consideration of the mutual influence of the fans by the LTC network link and the convolutional neural network link;
establishing a historical wind speed spatial data table according to the geographical position of the fan based on the historical wind speed data in the determined historical duration, taking the historical wind speed spatial data table as the input of the convolutional neural network link, and outputting prediction information considering fan mutual influence by the convolutional neural network link;
adding the prediction information considering the mutual influence of the fans and the prediction information considering the mutual influence of the fans, and then sequentially passing through two layers of fully-connected and corrected linear units to obtain a final predicted anemometer;
and calculating the frequency modulation energy of the wind power plant based on the wind speed of each fan in the final predicted anemometer at each moment.
Preferably, the constructing a historical wind speed time sequence data table based on the acquired historical wind speed data within the determined historical duration includes:
determining the wind speed value of each fan at each sampling moment in the determined historical time length according to the set wind speed sampling rate based on the acquired historical wind speed data in the determined historical time length;
and arranging the wind speed values of each fan at each sampling moment in the determined historical duration into a line according to the reverse sequence of the occurrence moments, and forming a historical wind speed time sequence data table by all the lines formed by the wind speed values of all the fans at each sampling moment in the determined historical duration.
Preferably, the step of using the historical wind speed time series data table as an input of an LTC network link, and outputting prediction information considering mutual influence of fans by the LTC network link includes:
and sequentially inputting the historical wind speed time sequence data table into an LTC1 network, an LTC2 network, a full connection layer FC1 and a correction linear unit in an LTC network link to obtain prediction information under consideration of mutual influence of fans.
Preferably, the step of constructing a historical wind speed spatial data table based on the obtained historical wind speed data and the geographical location information of the wind turbine within the determined historical time includes:
determining the size of a historical wind speed spatial data table based on the number of the fans;
determining a fan ranking of a historical wind speed spatial data table based on the geographical position of the fan;
determining the wind speed value of each fan at each sampling moment in the determined historical time length according to the set wind speed sampling rate based on the historical wind speed data in the determined historical time length;
and sequentially putting the wind speed values of the fans at each sampling moment in the determined historical duration into the historical wind speed spatial data table according to the fan ranking of the historical wind speed spatial data table to obtain the historical wind speed spatial data table.
Preferably, the taking the historical wind speed spatial data table as an input of the convolutional neural network link, and outputting the prediction information considering the mutual influence of the wind turbines by the convolutional neural network link includes:
and sequentially inputting the historical wind speed spatial data table into a convolutional layer Conv1, a maximum pooling layer MaxPool1, a convolutional layer Conv2, a maximum pooling MaxPool2, a full-link layer FC2, a full-link layer FC3 and a correction linear unit of a convolutional neural network link to obtain prediction information considering mutual influence of the fans.
Preferably, the adding the prediction information considering the mutual influence of the fans and the prediction information considering the mutual influence of the fans sequentially passes through two layers of fully-connected and modified linear units to obtain a final predicted anemometer, and the method includes:
adding the prediction information considering the mutual influence of the fans and the prediction information considering the mutual influence of the fans to output a final predicted wind speed through full connection FC4, full connection FC5 and a corrected Linear unit Linear;
wherein the final predicted wind speed comprises: and (4) the final wind speed predicted value of each fan at each sampling moment.
Preferably, the calculating the wind farm frequency modulation energy based on the wind speed of each wind turbine in the final predicted wind speed table at each moment includes:
calculating the power of each fan at each moment based on the wind speed of each fan in the final predicted wind speed table at each moment;
and summing the power of all the fans at the same moment to obtain the frequency modulation energy of the wind power plant consisting of all the fans at the moment.
Preferably, the power of each fan at each moment is calculated according to the following formula:
Figure DEST_PATH_IMAGE001
in the formula (I), the compound is shown in the specification,
Figure DEST_PATH_IMAGE002
is the power of one fan at a certain moment,
Figure DEST_PATH_IMAGE003
for the wind energy utilization coefficient of the ith fan,
Figure DEST_PATH_IMAGE004
in order to be the density of the air,Ris the radius of the impeller,
Figure DEST_PATH_IMAGE005
and the wind speed value of the ith fan at the jth moment is shown.
Preferably, the frequency modulation energy is calculated according to the following formula:
Figure DEST_PATH_IMAGE006
in the formula (I), the compound is shown in the specification,
Figure DEST_PATH_IMAGE007
for the frequency modulation energy of the wind power plant at a certain time,
Figure 253521DEST_PATH_IMAGE002
is the power of one fan at a certain moment.
In another aspect, the present application further provides a wind power frequency modulation energy prediction system, including:
the prediction module is used for respectively constructing a historical wind speed time sequence data table and a historical wind speed spatial data table based on the acquired historical wind speed data and the geographical position information of the fan in the determined historical duration, respectively taking the historical wind speed time sequence data table and the historical wind speed spatial data table as the input of an LTC network link and a convolutional neural network link, and respectively outputting prediction information under consideration of mutual influence of the fans and prediction information under consideration of mutual influence of the fans by the LTC network link and the convolutional neural network link;
the comprehensive prediction module is used for adding the prediction information considering the mutual influence of the fans and the prediction information considering the mutual influence of the fans, and then sequentially performing two-layer full-connection and linear unit correction to obtain a final predicted anemometer;
and the frequency modulation energy calculation module is used for calculating the frequency modulation energy of the wind power plant based on the wind speed of each fan in the final predicted anemometer at each moment.
Preferably, the prediction module comprises: a first prediction module and a second prediction module;
the first prediction module is used for constructing a historical wind speed time sequence data table based on the historical wind speed data with the determined historical duration, inputting the historical wind speed time sequence data table into the LTC network link based on the historical wind speed time sequence data table, and obtaining prediction information under consideration of mutual influence of fans;
and the second prediction module is used for determining a historical wind speed spatial data table based on the historical wind speed data with the determined historical duration and the geographical position information of the fan, and inputting the historical wind speed spatial data table into the convolutional neural network link to obtain prediction information considering mutual influence of the fan.
Preferably, the first prediction module comprises:
the wind speed value determining submodule is used for determining the wind speed value of each fan at each sampling moment in the determined historical duration according to the set wind speed sampling rate on the basis of the acquired historical wind speed data in the determined historical duration;
the anemometer construction submodule is used for arranging the wind speed values of all sampling moments of each fan in a certain historical duration into a line according to the reverse sequence of the occurrence moments, and all the lines formed by the wind speed values of all the fans in all the sampling moments in the certain historical duration form a historical wind speed time sequence data table;
and the data prediction submodule is used for sequentially inputting the historical wind speed time sequence data table into an LTC1 network, an LTC2 network, a full connection layer FC1 and a correction linear unit in an LTC network link to obtain prediction information under consideration of mutual influence of fans. Preferably, the frequency modulation energy calculating module includes:
the power calculation submodule is used for calculating the power of each fan at each moment based on the wind speed of each fan at each moment in the final predicted wind speed table;
and the energy calculation submodule is used for summing the power of all the fans at the same moment to obtain the frequency modulation energy of the wind power plant consisting of all the fans at the moment.
In yet another aspect, the present application further provides a computer device, including: one or more processors;
a processor for executing one or more programs;
when the one or more programs are executed by the one or more processors, a wind power frequency modulation energy prediction method as described above is implemented.
In still another aspect, the present application further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed, implements a wind power frequency modulation energy prediction method as described above.
Compared with the prior art, the invention has the beneficial effects that:
1. the invention provides a wind power frequency modulation energy prediction method, which comprises the following steps: respectively constructing a historical wind speed time sequence data table and a historical wind speed spatial data table based on the obtained historical wind speed data and the geographical position information of the fan within the determined historical duration, respectively taking the historical wind speed time sequence data table and the historical wind speed spatial data table as the input of an LTC network link and a convolutional neural network link, and respectively outputting prediction information under consideration of mutual influence of the fans and prediction information under consideration of the mutual influence of the fans by the LTC network link and the convolutional neural network link; establishing a historical wind speed spatial data table according to geographical position information of the fan based on historical wind speed data in the determined historical duration, taking the historical wind speed spatial data table as the input of the convolutional neural network link, and outputting prediction information considering mutual influence of the fans by the convolutional neural network link; adding the prediction information considering the mutual influence of the fans and the prediction information considering the mutual influence of the fans, and then sequentially passing through two layers of fully-connected and corrected linear units to obtain a final predicted anemometer; and calculating the frequency modulation energy of the wind power plant based on the wind speed of each fan in the final predicted anemometer at each moment. According to the technical scheme, the wind speed of the fan is predicted in a mode of combining the LTC network link and the convolutional neural network link, the effects of small equipment size, low test calculation cost and high precision are achieved, and the problems of large network model and high operation cost in the prior art are solved.
Drawings
FIG. 1 is a schematic diagram of steps of a wind power frequency modulation energy prediction method of the invention;
FIG. 2 is a detailed flow chart of the wind power frequency modulation energy prediction method of the invention.
Detailed Description
The invention provides a method for predicting wind power frequency modulation energy based on an LTC (Liquid Time-constant Networks) cyclic convolution network. The LTC is a novel efficient, energy-saving and flexible cyclic neural network designed by the fact that MIT is inspired by biological neurons, can learn in a training phase, can continuously adapt to a changing data stream to predict future behaviors, and is more suitable for predicting the wind speed of a fan affected by weather, climate, terrain and fan arrangement; in addition, the multi-scale filter of the convolutional neural network can further extract the multivariable spatial characteristics of the wind power plant, which influence the wind speed distribution due to the arrangement of the fans. And predicting the wind speed of each fan by combining the LTC + convolutional network, so that the integral frequency modulation energy of the wind power plant is obtained.
The LTC is a novel efficient, energy-saving and flexible cyclic neural network designed by the fact that MIT is inspired by biological neurons, can learn in a training phase, can continuously adapt to a changing data stream to predict future behaviors, and is more suitable for predicting the wind speed of a fan affected by weather, climate, terrain and fan arrangement; in addition, the multi-scale filter of the convolutional neural network can further extract the multivariable spatial characteristics of the wind power plant, which influence the wind speed distribution due to the arrangement of the fans. And predicting the wind speed of each fan by combining the LTC + convolutional network, so that the integral frequency modulation energy of the wind power plant is obtained.
Example 1:
the invention provides a wind power frequency modulation energy prediction method, which is shown in figure 1: the method comprises the following steps:
s1: respectively constructing a historical wind speed time sequence data table and a historical wind speed spatial data table based on the obtained historical wind speed data and the geographical position information of the fan within the determined historical duration, respectively taking the historical wind speed time sequence data table and the historical wind speed spatial data table as the input of an LTC network link and a convolutional neural network link, outputting prediction information under consideration of mutual influence of the fans by the LTC network link, and outputting prediction information under consideration of the mutual influence of the fans by the convolutional neural network link;
s2: adding the prediction information considering the mutual influence of the fans and the prediction information considering the mutual influence of the fans, and then sequentially passing through two layers of fully-connected and corrected linear units to obtain a final predicted anemometer;
s3: and calculating the frequency modulation energy of the wind power plant based on the wind speed of each fan in the final predicted anemometer at each moment.
The technical scheme of the invention is described in detail in the following with reference to fig. 2:
respectively constructing a historical wind speed time sequence data table and a historical wind speed spatial data table for the obtained historical wind speed data and the geographical position information of the fan in the determined historical duration in the S1, respectively taking the historical wind speed time sequence data table and the historical wind speed spatial data table as the input of an LTC network link and a convolutional neural network link, and respectively outputting prediction information under consideration of mutual influence of the fan and prediction information under consideration of the mutual influence of the fan by the LTC network link and the convolutional neural network link, specifically comprising:
s101, determining input of LTC network linkV in1。 V in1Is that
Figure DEST_PATH_IMAGE008
The ratio of vitamin to vitamin is,nthe number of the fans is equal to that of the fans,
Figure DEST_PATH_IMAGE009
the number of the historical wind speeds of each fan is determined by determining the historical time lengtht 0(s) and wind speed sampling rateT s(s) determining.
Figure DEST_PATH_IMAGE010
。
Figure DEST_PATH_IMAGE011
。v 11Before the current moment of the first fant 0The wind speed value of (1).v 12Before the current moment of the first fant 0- T sThe wind speed value of (1) and so onv Inws1Before the current moment of the first fan
Figure DEST_PATH_IMAGE012
The wind speed value of (1).
S103，V in1The link of the LTC network consisting of the LTC1 network, the LTC2 network, the full connection layer FC1 and the modified Linear unit Linear is under considerationPrediction information of mutual influence of fansV 1（
Figure DEST_PATH_IMAGE013
），
Figure DEST_PATH_IMAGE014
The number of the wind speeds predicted by each fan is determined according to the predicted wind speed durationt 1(s) and wind speed sampling rate
Figure DEST_PATH_IMAGE015
And (6) determining.
Figure DEST_PATH_IMAGE016
。
S102, determining input of the convolutional neural network linkV in2。 V in2Is that
Figure DEST_PATH_IMAGE017
And (5) maintaining. First, the size of the historical wind speed space data table is determined
Figure DEST_PATH_IMAGE018
If, ifmAnd if not, rounding up, and then determining the fan arrangement of the historical wind speed space data table. And performing fuzzy ranking according to the geographical position of the fan, namely, first up-north, down-south, and then left-west-east and right-east. For example, with 8 fans, WT5, WT1, WT3, WT2, WT6, WT8, WT7 and WT4 are arranged in the order of north, south, top and bottom, and then the results are obtained first
Figure DEST_PATH_IMAGE019
. Each row is further arranged by left, right and east, respectively, such as the first row, by left, right and east in order WT1, WT3, and WT 5. The second row, left, west, and right, is WT2, WT8, and WT6, respectively. The third row, left, right, and east, is WT7 and WT 4. The historical wind speed spatial data table is
Figure DEST_PATH_IMAGE020
Then, then
Figure DEST_PATH_IMAGE021
。
S104，V in2Obtaining prediction information considering mutual influence of fans through a convolutional neural network link consisting of a convolutional layer Conv1, a maximum pooling layer Maxpool1, a convolutional layer Conv2, a maximum pooling formation Maxpool2, a full connection layer FC2, a full connection layer FC3 and a correction Linear unit LinearV 2（
Figure 514869DEST_PATH_IMAGE013
）。
Adding the prediction information considering the mutual influence of the fans and the prediction information considering the mutual influence of the fans in the S2, and then sequentially performing two-layer full-connection and linear unit correction to obtain a final predicted anemometer, wherein the specific contents are as follows:
step five, mixingV 1AndV 2adding the output of the final predicted wind speed through the full-connection FC4, the full-connection FC5 and the corrected Linear unit Linear
Figure DEST_PATH_IMAGE022
（
Figure 953328DEST_PATH_IMAGE013
）；
Figure DEST_PATH_IMAGE023
。
S106, predicting the short-term wind speed of each fan at each moment
Figure DEST_PATH_IMAGE024
（i=1~n，j=1~outws）。
Calculating the wind power plant frequency modulation energy based on the wind speed of each fan in the final predicted wind speed table at each moment in S3, wherein the specific contents are as follows:
s107, calculating the available power of each fan at each moment
Figure DEST_PATH_IMAGE025
And wind farm frequency modulated energy
Figure DEST_PATH_IMAGE026
In the formula
Figure 925701DEST_PATH_IMAGE003
Is as followsiThe wind energy utilization coefficient of the platform fan,
Figure 5653DEST_PATH_IMAGE004
in order to be the density of the air,Ris the impeller radius.
Example 2
The invention based on the same invention concept also provides a wind power frequency modulation energy prediction system based on the LTC cyclic convolution network, which comprises the following steps:
the prediction module is used for respectively constructing a historical wind speed time sequence data table and a historical wind speed space data table based on the obtained historical wind speed data in the determined historical duration, respectively taking the historical wind speed time sequence data table and the historical wind speed space data table as the input of an LTC network link and a convolutional neural network link, and respectively outputting prediction information which considers the mutual influence of the fans by considering the prediction information of the mutual influence of the fans;
the comprehensive prediction module is used for adding the prediction information considering the mutual influence of the fans and the prediction information considering the mutual influence of the fans, and then sequentially performing two-layer full-connection and linear unit correction to obtain a final predicted anemometer;
and the frequency modulation energy calculation module is used for calculating the frequency modulation energy of the wind power plant based on the wind speed of each fan in the final predicted anemometer at each moment.
The prediction module comprises: a first prediction module and a second prediction module;
the first prediction module is used for constructing a historical wind speed time sequence data table based on the historical wind speed data with the determined historical duration, inputting the historical wind speed time sequence data table into the LTC network link based on the historical wind speed time sequence data table, and obtaining prediction information under consideration of mutual influence of fans;
and the second prediction module is used for determining a historical wind speed spatial data table based on the historical wind speed data with the determined historical duration and the acquired geographical position of the fan, and inputting the historical wind speed spatial data table into the convolutional neural network link to obtain prediction information considering mutual influence of the fan.
Preferably, the first prediction module comprises:
the wind speed value determining submodule is used for determining the wind speed value of each fan at each sampling moment in the determined historical duration according to the set wind speed sampling rate on the basis of the acquired historical wind speed data in the determined historical duration;
the anemometer construction submodule is used for arranging the wind speed values of all sampling moments of each fan in a certain historical duration into a line according to the reverse sequence of the occurrence moments, and all the lines formed by the wind speed values of all the fans in all the sampling moments in the certain historical duration form a historical wind speed time sequence data table;
and the data prediction submodule is used for sequentially inputting the historical wind speed time sequence data table into an LTC1 network, an LTC2 network, a full connection layer FC1 and a correction linear unit in an LTC network link to obtain prediction information under consideration of mutual influence of fans.
Preferably, the frequency modulation energy calculating module includes:
the power calculation submodule is used for calculating the power of each fan at each moment based on the wind speed of each fan at each moment in the final predicted wind speed table;
and the energy calculation submodule is used for summing the power of all the fans at the same moment to obtain the frequency modulation energy of the wind power plant consisting of all the fans at the moment.
The power of each fan at each moment is calculated according to the following formula:
Figure 906744DEST_PATH_IMAGE001
in the formula (I), the compound is shown in the specification,
Figure 439356DEST_PATH_IMAGE002
is the power of one fan at a certain moment,
Figure 952771DEST_PATH_IMAGE003
for the wind energy utilization coefficient of the ith fan,
Figure 949546DEST_PATH_IMAGE004
in order to be the density of the air,Ris the radius of the impeller,
Figure 474068DEST_PATH_IMAGE005
and the wind speed value of the ith fan at the jth moment is shown.
Preferably, the frequency modulation energy is calculated according to the following formula:
Figure 244709DEST_PATH_IMAGE006
in the formula (I), the compound is shown in the specification,
Figure 44038DEST_PATH_IMAGE007
for the frequency modulation energy of the wind power plant at a certain time,
Figure 144587DEST_PATH_IMAGE002
is the power of one fan at a certain moment.
The second prediction module specifically includes:
determining the size of a historical wind speed spatial data table based on the number of the fans;
determining a fan ranking of a historical wind speed spatial data table based on the geographical position of the fan;
determining the wind speed value of each fan at each sampling moment in the determined historical time length according to the set wind speed sampling rate based on the historical wind speed data in the determined historical time length;
sequentially placing the wind speed values of the fans at each sampling moment in a determined historical duration into the historical wind speed spatial data table according to the fan ranking of the historical wind speed spatial data table to obtain a historical wind speed spatial data table;
sequentially inputting the historical wind speed spatial data table into a convolutional layer Conv1, a maximum pooling layer MaxPool1, a convolutional layer Conv2, a maximum pooling formation MaxPool2, a full-link layer FC2, a full-link layer FC3 and a correction linear unit of a convolutional neural network link to obtain prediction information considering mutual influence of the fans;
the prediction information considering the mutual influence of the fans comprises the wind speed prediction value of each fan considering the mutual influence of the fans at each sampling moment.
The specific functions of the comprehensive prediction module comprise:
adding the prediction information considering the mutual influence of the fans and the prediction information considering the mutual influence of the fans to output a final predicted wind speed through full connection FC4, full connection FC5 and a corrected Linear unit Linear;
wherein the final predicted wind speed comprises: and (4) the final wind speed predicted value of each fan at each sampling moment.
For convenience of description, each part of the above apparatus is separately described as each module or unit by dividing the function. 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.
Based on the same inventive concept, in yet another embodiment of the present invention, a computer apparatus is provided, which includes a processor and a memory for storing a computer program comprising program instructions, and the processor is configured to execute the program instructions stored by the computer storage medium. The Processor may be a Central Processing Unit (CPU), or may be other general-purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable gate array (FPGA) or other Programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, etc., which is a computing core and a control core of the terminal, and is specifically adapted to load and execute one or more instructions in a computer storage medium to implement a corresponding method flow or a corresponding function; the processor provided by the embodiment of the invention can be used for executing the steps of the wind power frequency modulation energy prediction method.
Based on the same inventive concept, in yet another embodiment of the present invention, the present invention further provides a storage medium, specifically a computer-readable storage medium (Memory), which is a Memory device in a computer device and is used for storing programs and data. It is understood that the computer readable storage medium herein can include both built-in storage media in the computer device and, of course, extended storage media supported by the computer device. The computer-readable storage medium provides a storage space storing an operating system of the terminal. Also, one or more instructions, which may be one or more computer programs (including program code), are stored in the memory space and are adapted to be loaded and executed by the processor. It should be noted that the computer-readable storage medium may be a high-speed RAM memory, or may be a non-volatile memory (non-volatile memory), such as at least one disk memory. One or more instructions stored in the computer-readable storage medium may be loaded and executed by the processor to implement the corresponding steps of the wind power frequency modulation energy prediction method in the foregoing embodiments.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention 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 invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. 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 present invention is not limited to the above embodiments, and any modifications, equivalent substitutions, improvements, etc. within the spirit and principle of the present invention are included in the scope of the claims of the present invention.

Claims (11)

1. A wind power frequency modulation energy prediction method is characterized by comprising the following steps:
respectively constructing a historical wind speed time sequence data table and a historical wind speed spatial data table based on the obtained historical wind speed data and the geographical position information of the fan within the determined historical duration, respectively taking the historical wind speed time sequence data table and the historical wind speed spatial data table as the input of an LTC network link and a convolutional neural network link, and respectively outputting prediction information under consideration of mutual influence of the fans and prediction information under consideration of the mutual influence of the fans by the LTC network link and the convolutional neural network link;
adding the prediction information considering the mutual influence of the fans and the prediction information considering the mutual influence of the fans, and then sequentially passing through two layers of fully-connected and corrected linear units to obtain a final predicted anemometer;
calculating the frequency modulation energy of the wind power plant based on the wind speed of each fan in the final predicted anemometer at each moment;
the historical wind speed time sequence data table is constructed based on the obtained historical wind speed data in the determined historical duration, and the method comprises the following steps:
determining the wind speed value of each fan at each sampling moment in the determined historical time length according to the set wind speed sampling rate based on the acquired historical wind speed data in the determined historical time length;
arranging the wind speed values of each fan at each sampling moment in the determined historical duration into a line according to the reverse sequence of the occurrence moments, and forming a historical wind speed time sequence data table by all the lines formed by the wind speed values of all the fans at each sampling moment in the determined historical duration;
the historical wind speed spatial data table is constructed based on the obtained historical wind speed data and the geographical position information of the fan within the determined historical duration, and the method comprises the following steps:
determining the size of a historical wind speed spatial data table based on the number of the fans;
determining the fan ranking of a historical wind speed spatial data table based on the geographical position information of the fan;
determining the wind speed value of each fan at each sampling moment in the determined historical time length according to the set wind speed sampling rate based on the historical wind speed data in the determined historical time length;
and sequentially putting the wind speed values of the fans at each sampling moment in the determined historical duration into the historical wind speed spatial data table according to the fan ranking of the historical wind speed spatial data table to obtain the historical wind speed spatial data table.
2. The method of claim 1, wherein the using the historical wind speed timing data table as an input to an LTC network link, outputting, by the LTC network link, forecast information that underconsiders wind turbine interactions, comprises:
and sequentially inputting the historical wind speed time sequence data table into an LTC1 network, an LTC2 network, a full connection layer FC1 and a correction linear unit in an LTC network link to obtain prediction information under consideration of mutual influence of fans.
3. The method of claim 1, wherein the taking the historical wind speed spatial data table as an input to the convolutional neural network link, and outputting prediction information from the convolutional neural network link that considers wind turbine interactions, comprises:
and sequentially inputting the historical wind speed spatial data table into a convolutional layer Conv1, a maximum pooling layer MaxPool1, a convolutional layer Conv2, a maximum pooling MaxPool2, a full-link layer FC2, a full-link layer FC3 and a correction linear unit of a convolutional neural network link to obtain prediction information considering mutual influence of the fans.
4. The method of claim 1, wherein the adding the forecast information under consideration of the wind turbine interactions and the forecast information under consideration of the wind turbine interactions sequentially passes through two layers of fully connected and modified linear units to obtain a final forecast anemometer, comprising:
adding the prediction information considering the mutual influence of the fans and the prediction information considering the mutual influence of the fans to output a final predicted wind speed through full connection FC4, full connection FC5 and a corrected Linear unit Linear;
wherein the final predicted wind speed comprises: and (4) the final wind speed predicted value of each fan at each sampling moment.
5. The method of claim 1, wherein calculating wind farm frequency modulated energy based on wind speeds at each time of each wind turbine in the final predicted anemometer comprises:
calculating the power of each fan at each moment based on the wind speed of each fan in the final predicted wind speed table at each moment;
and summing the power of all the fans at the same moment to obtain the frequency modulation energy of the wind power plant consisting of all the fans at the moment.
6. The method of claim 5, wherein the power of each fan at each time is calculated as:
Figure 136512DEST_PATH_IMAGE001
in the formula (I), the compound is shown in the specification,
Figure 962166DEST_PATH_IMAGE002
is the power of one fan at a certain moment,
Figure 919758DEST_PATH_IMAGE003
for the wind energy utilization coefficient of the ith fan,
Figure 984666DEST_PATH_IMAGE004
in order to be the density of the air,Ris the radius of the impeller,
Figure 835947DEST_PATH_IMAGE005
and the wind speed value of the ith fan at the jth moment is shown.
7. The method of claim 5, wherein the frequency modulated energy is calculated as:
Figure 593687DEST_PATH_IMAGE006
in the formula (I), the compound is shown in the specification,
Figure 38575DEST_PATH_IMAGE007
for the frequency modulation energy of the wind power plant at a certain time,
Figure 143060DEST_PATH_IMAGE002
the power of one fan at a certain moment is shown, n is the number of the fans, and i is the number of the fans.
8. A wind power frequency modulation energy prediction system is characterized by comprising:
the prediction module is used for respectively constructing a historical wind speed time sequence data table and a historical wind speed spatial data table based on the acquired historical wind speed data and the geographical position information of the fan in the determined historical duration, respectively taking the historical wind speed time sequence data table and the historical wind speed spatial data table as the input of an LTC network link and a convolutional neural network link, and respectively outputting prediction information under consideration of mutual influence of the fans and prediction information under consideration of mutual influence of the fans by the LTC network link and the convolutional neural network link;
the comprehensive prediction module is used for adding the prediction information considering the mutual influence of the fans and the prediction information considering the mutual influence of the fans, and then sequentially performing two-layer full-connection and linear unit correction to obtain a final predicted anemometer;
the frequency modulation energy calculation module is used for calculating the frequency modulation energy of the wind power plant based on the wind speed of each fan in the final predicted anemometer at each moment;
the prediction module comprises: a first prediction module and a second prediction module;
the first prediction module is used for constructing a historical wind speed time sequence data table based on the historical wind speed data with the determined historical duration, inputting the historical wind speed time sequence data table into the LTC network link based on the historical wind speed time sequence data table, and obtaining prediction information under consideration of mutual influence of fans;
the second prediction module is used for determining a historical wind speed spatial data table based on the historical wind speed data with the determined historical duration and the acquired geographical position of the fan, and inputting the historical wind speed spatial data table into the convolutional neural network link to obtain prediction information considering mutual influence of the fan;
the first prediction module comprises:
the wind speed value determining submodule is used for determining the wind speed value of each fan at each sampling moment in the determined historical duration according to the set wind speed sampling rate on the basis of the acquired historical wind speed data in the determined historical duration;
the anemometer construction submodule is used for arranging the wind speed values of all sampling moments of each fan in a certain historical duration into a line according to the reverse sequence of the occurrence moments, and all the lines formed by the wind speed values of all the fans in all the sampling moments in the certain historical duration form a historical wind speed time sequence data table;
the data prediction submodule is used for sequentially inputting the historical wind speed time sequence data table into an LTC1 network, an LTC2 network, a full connection layer FC1 and a correction linear unit in an LTC network link to obtain prediction information under consideration of mutual influence of fans;
the second prediction module specifically includes:
determining the size of a historical wind speed spatial data table based on the number of the fans;
determining a fan ranking of a historical wind speed spatial data table based on the geographical position of the fan;
determining the wind speed value of each fan at each sampling moment in the determined historical time length according to the set wind speed sampling rate based on the historical wind speed data in the determined historical time length;
sequentially placing the wind speed values of the fans at each sampling moment in a determined historical duration into the historical wind speed spatial data table according to the fan ranking of the historical wind speed spatial data table to obtain a historical wind speed spatial data table;
sequentially inputting the historical wind speed spatial data table into a convolutional layer Conv1, a maximum pooling layer MaxPool1, a convolutional layer Conv2, a maximum pooling formation MaxPool2, a full-link layer FC2, a full-link layer FC3 and a correction linear unit of a convolutional neural network link to obtain prediction information considering mutual influence of the fans;
the prediction information considering the mutual influence of the fans comprises the wind speed prediction value of each fan considering the mutual influence of the fans at each sampling moment.
9. The system of claim 8, wherein the fm energy calculation module comprises:
the power calculation submodule is used for calculating the power of each fan at each moment based on the wind speed of each fan at each moment in the final predicted wind speed table;
and the energy calculation submodule is used for summing the power of all the fans at the same moment to obtain the frequency modulation energy of the wind power plant consisting of all the fans at the moment.
10. A computer device, comprising: one or more processors;
a processor for executing one or more programs;
the one or more programs, when executed by the one or more processors, implement a wind farm modulated frequency energy prediction method as recited in any of claims 1-7.
11. A computer-readable storage medium having stored thereon a computer program which, when executed, implements a wind power frequency modulation energy prediction method as claimed in any one of claims 1 to 7.
CN202111398040.2A 2021-11-24 2021-11-24 Wind power frequency modulation energy prediction method and system and computer equipment Active CN113824115B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111398040.2A CN113824115B (en) 2021-11-24 2021-11-24 Wind power frequency modulation energy prediction method and system and computer equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111398040.2A CN113824115B (en) 2021-11-24 2021-11-24 Wind power frequency modulation energy prediction method and system and computer equipment

Publications (2)

Publication Number Publication Date
CN113824115A CN113824115A (en) 2021-12-21
CN113824115B true CN113824115B (en) 2022-03-08

Family

ID=78919771

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111398040.2A Active CN113824115B (en) 2021-11-24 2021-11-24 Wind power frequency modulation energy prediction method and system and computer equipment

Country Status (1)

Country Link
CN (1) CN113824115B (en)

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP5342496B2 (en) * 2010-04-16 2013-11-13 株式会社日立製作所 Wind turbine generator group control device and control method thereof
CN110334589B (en) * 2019-05-23 2021-05-14 中国地质大学（武汉） High-time-sequence 3D neural network action identification method based on hole convolution
CN111784041B (en) * 2020-06-28 2024-02-09 中国电力科学研究院有限公司 Wind power prediction method and system based on graph convolution neural network
CN112529282A (en) * 2020-12-03 2021-03-19 清华大学 Wind power plant cluster short-term power prediction method based on space-time graph convolutional neural network
CN113642254A (en) * 2021-09-01 2021-11-12 中国人民解放军国防科技大学 Wind power generation power prediction method based on multi-scale attention mechanism

Also Published As

Publication number Publication date
CN113824115A (en) 2021-12-21

Similar Documents

Publication Publication Date Title
CN112348292B (en) Short-term wind power prediction method and system based on deep learning network
CN112217202B (en) Distributed new energy, energy storage and power distribution network planning method considering flexibility investment
CN113937763B (en) Wind power prediction method, device, equipment and storage medium
CN103106544B (en) A kind of photovoltaic generation prognoses system based on T-S Fuzzy neutral net
CN107392304A (en) A kind of Wind turbines disorder data recognition method and device
CN114006369B (en) Regional wind and light station power joint prediction method and device, electronic equipment and storage medium
CN104732300B (en) A kind of neutral net wind power short term prediction method theoretical based on Fuzzy divide
CN107705556A (en) A kind of traffic flow forecasting method combined based on SVMs and BP neural network
CN105590144A (en) Wind speed prediction method and apparatus based on NARX neural network
CN110094300A (en) Wind turbines to windage losses antidote, device, equipment and medium
CN116599050A (en) Photovoltaic prediction method and related device based on self-attention mechanism
CN112417734A (en) Wind speed correction method and device based on geographic information of wind power plant
CN116799796A (en) Photovoltaic power generation power prediction method, device, equipment and medium
CN113824115B (en) Wind power frequency modulation energy prediction method and system and computer equipment
CN113673788A (en) Photovoltaic power generation power prediction method based on decomposition error correction and deep learning
CN107527093B (en) Wind turbine generator running state diagnosis method and device
CN117371581A (en) New energy generated power prediction method, device and storage medium
CN113139671A (en) Power load interval prediction method and system based on branch structure model
Wang et al. Construction of offline predictive controller for wind farm based on CNN–GRNN
CN117094119A (en) Reservoir dispatching method and computer taking power generation, ecological flow and surface water temperature into consideration
CN114330649B (en) Voltage regulation method and system based on evolutionary learning and deep reinforcement learning
CN113224788A (en) Static voltage stability evaluation method for AC/DC hybrid power grid with wind power access
CN111651939A (en) Dynamic equivalent modeling method for permanent magnet wind power plant considering converter control parameter difference
Lim et al. R&D Transitions in Response to Digital Transformation in Korea
CN117117858A (en) Wind turbine generator power prediction method, device and storage medium

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