CN115034422A - Wind power short-term power prediction method and system based on fluctuation identification and error correction - Google Patents
Wind power short-term power prediction method and system based on fluctuation identification and error correction Download PDFInfo
- Publication number
- CN115034422A CN115034422A CN202110240352.4A CN202110240352A CN115034422A CN 115034422 A CN115034422 A CN 115034422A CN 202110240352 A CN202110240352 A CN 202110240352A CN 115034422 A CN115034422 A CN 115034422A
- Authority
- CN
- China
- Prior art keywords
- wind speed
- fluctuation
- power
- wind
- low
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 93
- 238000012937 correction Methods 0.000 title claims abstract description 74
- 238000012545 processing Methods 0.000 claims description 30
- 238000012549 training Methods 0.000 claims description 15
- 238000013528 artificial neural network Methods 0.000 claims description 14
- 238000000354 decomposition reaction Methods 0.000 claims description 12
- 238000007781 pre-processing Methods 0.000 claims description 8
- 230000001174 ascending effect Effects 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 claims description 5
- 230000001419 dependent effect Effects 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 7
- 238000010586 diagram Methods 0.000 description 11
- 238000004590 computer program Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 238000010248 power generation Methods 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000010521 absorption reaction Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000000053 physical method Methods 0.000 description 1
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
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
-
- 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
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Primary Health Care (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Wind Motors (AREA)
Abstract
The invention relates to a wind power short-term power prediction method and a system based on fluctuation identification and error correction, comprising the following steps: dividing the wind speed waveform of the duration to be predicted into a plurality of wind speed fluctuations, and forming a characteristic sequence of each wind speed fluctuation; obtaining low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation by utilizing a pre-established wind power short-term power prediction model and the characteristic sequence of the wind speed fluctuation; and generating wind power fluctuation corresponding to each wind speed fluctuation according to the low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation. The technical scheme provided by the invention breaks through the limitation that the effect of capturing wind power output fluctuation cannot be achieved or is limited by the conventional short-term power prediction method, and improves the accuracy of the wind power short-term power.
Description
Technical Field
The invention relates to the field of new energy power generation in a power system, in particular to a wind power short-term power prediction method and system based on fluctuation identification and error correction.
Background
Because the wind power accumulated installed capacity is increased year by year, a power grid dispatching department needs to make a dispatching plan according to a wind power forecasting result so as to promote the consumption of wind power generation with randomness and volatility, and therefore a wind power forecasting system needs to be deployed on both a power grid dispatching side and a wind farm side.
With the development of many years, the accuracy and the application degree of wind power prediction are gradually improved, but the current prediction level cannot meet the actual requirement of scheduling operation. In the technical aspect, the currently applied relatively mature algorithms include a physical method, a statistical method and a combination method, but from the basic principle, the methods all use a single point as a prediction target, namely, a meteorological-power association relation is established based on historical operating data, power at a corresponding moment is predicted by using meteorological prediction data moment by moment in a future time period, the method does not consider the time sequence characteristics of a wind power sequence, the fluctuation of the wind power cannot be effectively predicted, and the prediction precision reaches the bottleneck.
A typical example, a BPNN-based wind power short-term power prediction method, is listed here, and the method is divided into two stages, namely model training and prediction execution. In the model training stage, a BPNN is utilized to establish a correlation model of the power and the meteorological elements related to the numerical weather forecast by collecting historical operation data (actual power, numerical weather forecast data and the like) of the wind power plant; in the prediction execution stage, numerical weather forecast data is used as input, the forecast values of the relevant weather elements are input into a prediction model, and the predicted power of the wind power plant in a certain period of time in the future is obtained.
As can be seen from fig. 1, in the BPNN-based wind power short-term power prediction method, in the training and prediction stages, a single moment is used as an object, a relation between weather and power at a corresponding moment is established, and a power prediction value at the corresponding moment is obtained based on weather prediction data at different moments in the future, and the time sequence relevance of wind power output is not basically considered, so that the volatility of the output cannot be effectively tracked by a prediction result, and the maximum and minimum values of the wind power output are difficult to accurately predict.
At present, the wind power plant short-term power prediction result oriented to the fluctuation process is published publicly, and the technical route of the prediction result is as follows: in the training stage, power is divided into low-frequency components and high-frequency components by frequency division of historical power based on historical power of a wind power plant and weather forecast data; for low-frequency components, adopting a prediction method taking a fluctuation process as an object, dividing the fluctuation into different fluctuation types, extracting the characteristics of different types of fluctuation, and respectively establishing prediction models; for the high-frequency component, modeling by adopting a BPNN method; in the prediction execution stage, a numerical weather forecast is used as input, forecast data of the wind speed is extracted, the wind speed is filtered, the low-frequency component of the wind speed is used as input, fluctuation characteristics of the low-frequency component are extracted, and the low-frequency component is input into a prediction model of a corresponding type, so that a low-frequency component prediction value of the power is obtained. And directly inputting the original wind speed sequence into the BPNN model to obtain a high-frequency component predicted value of the power, and superposing the low frequency and the high frequency to obtain a final predicted value of the power.
Through the analysis of actual data, the fluctuation of wind power cannot be effectively tracked in the short-term prediction scale of the numerical weather forecast, the low-frequency component and the high-frequency component of the power are respectively predicted and superposed in practical application, the high-frequency component is basically predicted in an invalid manner, because the correlation between a wind speed sequence and the high-frequency component is very low, the statistical rules of the wind speed sequence and the high-frequency component cannot be basically obtained, and the high-frequency component predicted by adopting a BPNN (Business process neural network) manner can be an erroneous interference item for the final prediction result.
In view of the above, there is a need for a wind power short-term power prediction method that can overcome the above-mentioned drawbacks.
Disclosure of Invention
Aiming at the defects of the prior art, the invention aims to provide a wind power short-term power prediction method and a wind power short-term power prediction system based on fluctuation identification and error correction, the method breaks through the limitation that the effect of the conventional short-term power prediction method that the wind power output fluctuation cannot be captured or is limited, and the accuracy of the wind power short-term power is improved.
The purpose of the invention is realized by adopting the following technical scheme:
the invention provides a wind power short-term power prediction method based on fluctuation identification and error correction, which comprises the following steps:
dividing the wind speed waveform of the duration to be predicted into a plurality of wind speed fluctuations, and forming a characteristic sequence of each wind speed fluctuation;
obtaining low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation by utilizing a pre-established wind power short-term power prediction model and the characteristic sequence of the wind speed fluctuation;
generating wind power fluctuation corresponding to each wind speed fluctuation according to the low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation;
and the wind speed waveform of the duration to be predicted is generated based on wind speed forecast data at each moment in the duration to be predicted.
Preferably, the training process of the pre-established wind power short-term power prediction model includes:
dividing the wind speed waveform of the historical duration into a plurality of wind speed fluctuations, and forming a characteristic sequence of each wind speed fluctuation;
preprocessing the actual wind power waveform with the historical duration to obtain each low-frequency power fluctuation corresponding to each wind speed fluctuation time divided by the wind speed waveform with the historical duration;
uniformly processing the low-frequency power fluctuation into low-frequency power fluctuation with the same time length by using a cubic spline interpolation method;
and as input data of the initial BPNN neural network, taking each low-frequency power fluctuation after interpolation as output data of the initial BPNN neural network, training the initial BPNN neural network, and obtaining a wind power short-term power prediction model.
Preferably, the determination of the wind speed fluctuation comprises:
marking a maximum value and a minimum value in the wind speed waveform;
and dividing the wind speed waveform into a plurality of wind speed fluctuations according to the principle that the adjacent waveform segment corresponding to the minimum value-maximum value-minimum value is a fluctuation.
Further, the preprocessing the actual wind power waveform of the historical duration to obtain each low-frequency power fluctuation corresponding to each wind speed fluctuation time divided by the wind speed waveform of the historical duration includes:
performing wavelet decomposition on the actual wind power waveform with the historical duration by adopting a db6 wavelet decomposition technology to obtain each layer of profile coefficients;
reconstructing the profile coefficients of each layer to obtain low-frequency power waveforms corresponding to the profile coefficients of each layer;
respectively calculating Pearson correlation coefficients between low-frequency power waveforms corresponding to the profile coefficients of each layer and actual wind power waveforms of historical duration, and selecting a low-frequency power waveform with the highest correlation coefficient;
and intercepting the selected low-frequency power waveform to obtain each low-frequency power fluctuation corresponding to each wind speed fluctuation time divided by the wind speed waveform of the historical duration.
Preferably, the generating wind power fluctuation corresponding to each wind speed fluctuation according to the low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation includes:
carrying out reverse interpolation processing on low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation;
and correcting the power value in each low-frequency power fluctuation after the reverse interpolation processing to obtain the wind power fluctuation corresponding to each wind speed fluctuation.
Further, the performing of reverse interpolation processing on the low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation includes:
and (3) reversely interpolating the low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation into low-frequency power fluctuation consistent with each wind speed fluctuation time length by adopting a cubic spline interpolation method, and adjusting the time node of the low-frequency power fluctuation to be consistent with the corresponding wind speed fluctuation.
Further, the correcting the power value in each low-frequency power fluctuation after the reverse interpolation includes:
determining a cluster to which the wind speed fluctuation belongs based on a characteristic sequence of the wind speed fluctuation corresponding to the low-frequency power fluctuation after reverse interpolation processing;
calling a first correction coefficient and a second correction coefficient corresponding to the cluster;
substituting the first correction coefficient and the second correction coefficient corresponding to the cluster and the power value in the low-frequency power fluctuation into a correction equation to obtain a correction value of the power value in the low-frequency power fluctuation;
wherein the clustering is based on clustering of individual wind speed fluctuations demarcated from historical wind speed waveforms.
Further, the process of determining the first correction coefficient and the second correction coefficient corresponding to the cluster includes:
substituting the characteristic sequence of the wind speed fluctuation contained in the cluster into a pre-established wind power short-term power prediction model to obtain low-frequency power fluctuation corresponding to the wind speed fluctuation contained in the cluster;
carrying out reverse interpolation processing on low-frequency power fluctuation corresponding to wind speed fluctuation contained in the clusters;
and taking the actual wind power at each moment in the wind speed fluctuation contained in the cluster as a dependent variable in a correction equation, taking the power value at the corresponding moment in the low-frequency power fluctuation corresponding to the wind speed fluctuation contained in the cluster after reverse interpolation processing as an independent variable in the correction equation, and fitting by adopting a linear least square method to obtain a first correction coefficient and a second correction coefficient corresponding to the cluster.
Further, the correction equation is:
the corrected value of the power value is A multiplied by the power value + B;
wherein A is a first fitting coefficient, and B is a second fitting coefficient.
Preferably, the characteristics in the characteristic sequence of the wind speed fluctuation include, but are not limited to:
minimum value, maximum value, time length between two minimum values in wind speed fluctuation, ratio of slope of wind speed ascending section and wind speed descending section, mean value and standard deviation.
The invention provides a wind power short-term power prediction system based on fluctuation identification and error correction, which comprises:
the device comprises a dividing module, a time-length predicting module and a time-length predicting module, wherein the dividing module is used for dividing a wind speed waveform of a time length to be predicted into a plurality of wind speed fluctuations and forming a characteristic sequence of each wind speed fluctuation;
the acquisition module is used for acquiring low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation by utilizing a pre-established wind power short-term power prediction model and the characteristic sequence of the wind speed fluctuation;
the generating module is used for generating wind power fluctuation corresponding to each wind speed fluctuation according to the low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation;
and the wind speed waveform of the duration to be predicted is generated based on wind speed forecast data at each moment in the duration to be predicted.
Compared with the closest prior art, the invention has the following beneficial effects:
according to the technical scheme provided by the invention, the wind speed waveform of the duration to be predicted is divided into a plurality of wind speed fluctuations, and a characteristic sequence of each wind speed fluctuation is formed; obtaining low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation by utilizing a pre-established wind power short-term power prediction model and the characteristic sequence of the wind speed fluctuation; and generating wind power fluctuation corresponding to each wind speed fluctuation according to the low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation. The scheme breaks through the limitation that the effect of the conventional short-term power prediction method for capturing or catching the wind power output fluctuation is limited, and improves the accuracy of the wind power short-term power.
Drawings
FIG. 1 is a schematic diagram of wind power short-term power prediction effect based on BPNN;
FIG. 2 is a flow chart of a wind power short-term power prediction method based on fluctuation identification and error correction;
FIG. 3 is a structural diagram of a wind power short-term power prediction system based on fluctuation identification and error correction;
FIG. 4 is a schematic diagram for comparing the prediction effects of the prediction methods of different wind power short-term powers in the embodiment of the present invention.
Detailed Description
The following provides a more detailed description of embodiments of the present invention, with reference to the accompanying drawings.
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1:
the method for predicting the fluctuation process of the wind power is an important means for breaking through the bottleneck of the existing prediction method and improving the prediction precision. Aiming at the problems that the continuity and the volatility of wind power output are not considered in the existing mature wind power short-term prediction method, the output fluctuation process of wind power is difficult to effectively capture, and the published prediction for the fluctuation process is difficult to solve, the invention provides a wind power short-term power prediction method based on fluctuation identification and error correction, as shown in figure 2, the method breaks through the limitation that the existing short-term power prediction method cannot capture or capture the wind power output fluctuation with limited effect, can improve the accuracy of the wind power short-term power, and comprehensively serves the scheduling operation and the high-efficiency absorption of the wind power.
The method comprises the following specific steps:
step 101, dividing a wind speed waveform of a duration to be predicted into a plurality of wind speed fluctuations, and forming a characteristic sequence of each wind speed fluctuation;
102, obtaining low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation by utilizing a pre-established wind power short-term power prediction model and the characteristic sequence of the wind speed fluctuation;
103, generating wind power fluctuation corresponding to each wind speed fluctuation according to the low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation;
the wind speed waveform of the time length to be predicted is generated based on wind speed forecast data at each moment in the time length to be predicted.
In the best embodiment of the invention, the wind power fluctuations corresponding to the wind speed fluctuations are combined according to time to obtain the wind power waveform corresponding to the duration to be predicted.
Specifically, the training process of the pre-established wind power short-term power prediction model includes:
dividing the wind speed waveform of the historical duration into a plurality of wind speed fluctuations, and forming a characteristic sequence of each wind speed fluctuation;
preprocessing the actual wind power waveform with the historical duration to obtain each low-frequency power fluctuation corresponding to each wind speed fluctuation time divided by the wind speed waveform with the historical duration;
uniformly processing the low-frequency power fluctuation into low-frequency power fluctuation with the same time length by using a cubic spline interpolation method;
and as input data of the initial BPNN neural network, taking each low-frequency power fluctuation after interpolation as output data of the initial BPNN neural network, and training the initial BPNN neural network to obtain a wind power short-term power prediction model.
Specifically, the determination of the wind speed fluctuation includes:
marking a maximum value and a minimum value in the wind speed waveform;
and dividing the wind speed waveform into a plurality of wind speed fluctuations according to the principle that the adjacent waveform segment corresponding to the minimum value-maximum value-minimum value is a fluctuation.
Specifically, the step 102 includes:
102-1, performing wavelet decomposition on the actual wind power waveform of the historical duration by adopting a db6 wavelet decomposition technology to obtain each layer of profile coefficients;
102-2, reconstructing each layer of profile coefficient to obtain a low-frequency power waveform corresponding to each layer of profile coefficient;
102-3, respectively calculating a Pearson correlation coefficient between a low-frequency power waveform corresponding to each layer of profile coefficient and an actual wind power waveform of historical duration, and selecting the low-frequency power waveform with the highest correlation coefficient;
and 102-4, intercepting the selected low-frequency power waveform to obtain each low-frequency power fluctuation corresponding to each wind speed fluctuation time divided by the wind speed waveform with the historical duration.
Specifically, the step 103 includes:
103-1, carrying out reverse interpolation processing on low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation;
and 103-2, correcting the power value in each low-frequency power fluctuation after the reverse interpolation processing to obtain wind power fluctuation corresponding to each wind speed fluctuation.
In the best embodiment of the invention, the high-frequency component of the wind power is taken into consideration by a correction mode, so that the accuracy of prediction is improved.
Further, the step 103-1 includes:
and (3) reversely interpolating the low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation into low-frequency power fluctuation consistent with the time length of each wind speed fluctuation by adopting a cubic spline interpolation method, and adjusting the time node of the low-frequency power fluctuation to be consistent with the corresponding wind speed fluctuation.
Further, the step 102-4 includes:
102-4-1, determining a cluster to which the wind speed fluctuation belongs based on a characteristic sequence of the wind speed fluctuation corresponding to the low-frequency power fluctuation after reverse interpolation processing;
step 102-4-2, calling a first correction coefficient and a second correction coefficient corresponding to the cluster;
step 102-4-3, substituting the first correction coefficient and the second correction coefficient corresponding to the cluster and the power value in the low-frequency power fluctuation into a correction equation to obtain a correction value of the power value in the low-frequency power fluctuation;
wherein the clustering is based on clustering of individual wind speed fluctuations demarcated from historical wind speed waveforms.
Still further, the determining process of the first correction coefficient and the second correction coefficient corresponding to the cluster includes:
substituting the characteristic sequence of the wind speed fluctuation contained in the cluster into a pre-established wind power short-term power prediction model to obtain low-frequency power fluctuation corresponding to the wind speed fluctuation contained in the cluster;
carrying out reverse interpolation processing on low-frequency power fluctuation corresponding to wind speed fluctuation contained in the clusters;
and taking the actual wind power at each moment in the wind speed fluctuation contained in the cluster as a dependent variable in a correction equation, taking the power value at the corresponding moment in the low-frequency power fluctuation corresponding to the wind speed fluctuation contained in the cluster after reverse interpolation processing as an independent variable in the correction equation, and fitting by adopting a linear least square method to obtain a first correction coefficient and a second correction coefficient corresponding to the cluster.
Still further, the correction equation is:
the corrected value of the power value is A multiplied by the power value + B;
wherein A is a first fitting coefficient and B is a second fitting coefficient.
Specifically, the characteristics in the characteristic sequence of the wind speed fluctuation include, but are not limited to:
minimum value, maximum value, time length between two minimum values in wind speed fluctuation, ratio of slope of wind speed ascending section to slope of wind speed descending section, average value and standard deviation.
Example 2:
the invention provides a wind power short-term power prediction system based on fluctuation identification and error correction, as shown in fig. 3, comprising:
the device comprises a dividing module, a time-length predicting module and a time-length predicting module, wherein the dividing module is used for dividing a wind speed waveform of a time length to be predicted into a plurality of wind speed fluctuations and forming a characteristic sequence of each wind speed fluctuation;
the acquisition module is used for acquiring low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation by utilizing a pre-established wind power short-term power prediction model and the characteristic sequence of the wind speed fluctuation;
the generating module is used for generating wind power fluctuation corresponding to each wind speed fluctuation according to the low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation;
and the wind speed waveform of the duration to be predicted is generated based on wind speed forecast data at each moment in the duration to be predicted.
Specifically, the system further includes: the modeling module is used for establishing a wind power short-term power prediction model in advance, and comprises:
the dividing unit is used for dividing the wind speed waveform of the historical duration into a plurality of wind speed fluctuations and forming a characteristic sequence of each wind speed fluctuation;
the preprocessing unit is used for preprocessing the actual wind power waveform with the historical duration to obtain each low-frequency power fluctuation corresponding to each wind speed fluctuation time divided by the wind speed waveform with the historical duration;
the interpolation unit is used for uniformly processing the low-frequency power fluctuation into low-frequency power fluctuation with the same time length by using a cubic spline interpolation method;
and the training unit is used for serving as input data of the initial BPNN neural network, taking each low-frequency power fluctuation after interpolation as output data of the initial BPNN neural network, training the initial BPNN neural network, and obtaining the wind power short-term power prediction model.
Specifically, the determination of the wind speed fluctuation includes:
marking a maximum value and a minimum value in the wind speed waveform;
and according to the principle that the adjacent waveform segment corresponding to the minimum value-maximum value-minimum value is a fluctuation, dividing the wind speed waveform into a plurality of wind speed fluctuations.
Further, the preprocessing unit includes:
the wavelet decomposition submodule is used for performing wavelet decomposition on the actual wind power waveform with the historical duration by adopting a db6 wavelet decomposition technology to obtain each layer of profile coefficients;
the wavelet reconstruction submodule is used for reconstructing each layer of profile coefficient to obtain a low-frequency power waveform corresponding to each layer of profile coefficient;
the selection submodule is used for respectively calculating a Pearson correlation coefficient between a low-frequency power waveform corresponding to each layer of profile coefficient and an actual wind power waveform of historical duration, and selecting a low-frequency power waveform with the highest correlation coefficient;
and the intercepting submodule is used for intercepting the selected low-frequency power waveform to obtain each low-frequency power fluctuation corresponding to each wind speed fluctuation time divided by the wind speed waveform with the historical duration.
Specifically, the obtaining module includes:
the reverse interpolation unit is used for performing reverse interpolation processing on low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation;
and the correcting unit is used for correcting the power value in each low-frequency power fluctuation after the reverse interpolation processing to obtain the wind power fluctuation corresponding to each wind speed fluctuation.
Further, the inverse interpolation unit is configured to:
and (3) reversely interpolating the low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation into low-frequency power fluctuation consistent with the time length of each wind speed fluctuation by adopting a cubic spline interpolation method, and adjusting the time node of the low-frequency power fluctuation to be consistent with the corresponding wind speed fluctuation.
Further, the correction unit includes:
the attribution determining submodule is used for determining a cluster to which the wind speed fluctuation belongs based on a characteristic sequence of the wind speed fluctuation corresponding to the low-frequency power fluctuation after reverse interpolation processing;
the calling submodule is used for calling a first correction coefficient and a second correction coefficient corresponding to the cluster;
the correction submodule is used for substituting the first correction coefficient and the second correction coefficient corresponding to the cluster and the power value in the low-frequency power fluctuation into a correction equation to obtain a correction value of the power value in the low-frequency power fluctuation;
wherein the clustering is based on clustering of individual wind speed fluctuations demarcated from historical wind speed waveforms.
Further, the modification unit further includes a coefficient determination submodule configured to determine a first modification coefficient and a second modification coefficient corresponding to a cluster, where the coefficient determination submodule includes:
the acquiring subunit is used for substituting the characteristic sequence of the wind speed fluctuation contained in the cluster into a pre-established wind power short-term power prediction model to obtain low-frequency power fluctuation corresponding to the wind speed fluctuation contained in the cluster;
a reverse interpolation subunit, configured to perform reverse interpolation processing on low-frequency power fluctuations corresponding to the wind speed fluctuations included in the cluster;
and the fitting subunit is used for taking the actual wind power at each moment in the wind speed fluctuation contained in the cluster as a dependent variable in the correction equation, taking the power value at the corresponding moment in the low-frequency power fluctuation corresponding to the wind speed fluctuation contained in the cluster after reverse interpolation processing as an independent variable in the correction equation, and fitting by adopting a linear least square method to obtain a first correction coefficient and a second correction coefficient corresponding to the cluster.
Further, the correction equation is:
the corrected value of the power value is A multiplied by the power value + B;
wherein A is a first fitting coefficient and B is a second fitting coefficient.
Specifically, the characteristics in the characteristic sequence of the wind speed fluctuation include, but are not limited to:
minimum value, maximum value, time length between two minimum values in wind speed fluctuation, ratio of slope of wind speed ascending section and wind speed descending section, mean value and standard deviation.
Example 3:
according to the wind power short-term power prediction method based on fluctuation identification and error correction, the fluctuation rules of wind power are analyzed, different types of fluctuation processes are established based on a clustering means, prediction models are respectively established for the different types of fluctuation processes, the fluctuation type corresponding to meteorological elements is used as input, the characteristics of the fluctuation processes are used as output, power reconstruction and error correction are carried out based on the characteristics, the fluctuation of the wind power can be effectively captured, the bottleneck that single-point prediction can reach precision is broken through, and the problem that the fluctuation of the wind power is difficult to accurately predict is solved;
in order to verify the accuracy of the effect, the wind power in the future half ten days of a certain wind power plant is predicted by adopting a conventional wind power prediction method and the prediction method on a simulation program;
the prediction method comprises the following specific steps:
step 1: generating a wind speed waveform corresponding to the period to be predicted by using wind speed forecast data at each moment in the period to be predicted;
and 2, step: marking a maximum value and a minimum value in the wind speed waveform corresponding to the period to be predicted, taking a waveform segment corresponding to the adjacent 'minimum value-maximum value-minimum value' as a wind speed fluctuation, and dividing the wind speed waveform corresponding to the period to be predicted to generate a plurality of wind speed fluctuations;
and 3, step 3: extracting features of each wind speed fluctuation, the features including: minimum value, maximum value, time length between two minimum values in wind speed fluctuation, ratio of slope of wind speed ascending section and wind speed descending section, average value and standard deviation.
And 4, step 4: substituting the characteristics of each wind speed fluctuation into a pre-established wind power prediction model to obtain the power fluctuation which is output by the model and corresponds to each wind speed fluctuation;
and 5: reconstructing power fluctuation corresponding to each wind speed fluctuation by utilizing a cubic spline interpolation method; and the time length of the power fluctuation after reconstruction is consistent with the time length of the corresponding wind speed fluctuation.
Step 6: judging the cluster to which each wind speed fluctuation belongs based on the characteristics of each wind speed fluctuation;
step 6-1: obtaining the characteristics based on the historical wind speed fluctuation to cluster the historical wind speed fluctuation, and obtaining clusters and cluster centers of the clusters;
step 6-2: calculating the distance from each fluctuation to the clustering center of each cluster based on the characteristics of each wind speed fluctuation;
and 6-3: for each wind speed fluctuation, the cluster corresponding to the cluster center whose distance is the smallest is taken as the cluster to which the wind speed fluctuation belongs.
And 7: correcting the reconstructed power fluctuation corresponding to each wind speed fluctuation by using the fitting coefficients A and B corresponding to the cluster to which each wind speed fluctuation belongs; the correction method comprises the following steps: and the actual wind power is A multiplied by the predicted wind power + B.
And 8: and recombining all the corrected power fluctuations to generate a power waveform corresponding to the period to be predicted.
The pre-established training process of the wind power prediction model comprises the following steps:
step A: and obtaining a wind speed waveform corresponding to the historical period and a power waveform with the highest similarity to the wind speed waveform based on the wind speed forecast data and the actual wind power data at each moment in the historical period.
A-1: acquiring wind speed forecast data and actual wind power data at each moment in a historical period, and rejecting unreasonable data;
a-2: constructing a wind speed waveform and an actual wind power waveform corresponding to a historical period by using wind speed forecast data and actual wind power data at each moment in the historical period;
a-3: performing K-layer wavelet decomposition on an actual wind power waveform corresponding to a historical period based on a multi-scale one-dimensional wavelet decomposition means to obtain K-layer profile coefficients; for example: performing 5-layer frequency decomposition on the actual wind power waveform corresponding to the historical period by using a 'db 6' wavelet to obtain 5-layer profile coefficients;
a-4: reconstructing the general picture coefficient of the K layers to obtain K power waveforms;
a-5: and respectively calculating Pearson correlation coefficients of the K power waveforms and the wind speed waveforms corresponding to the historical period, and selecting the power waveform with the highest correlation coefficient as the power waveform with the highest wind speed waveform similarity corresponding to the historical period.
And B: dividing the wind speed waveform corresponding to the historical period into a plurality of historical wind speed fluctuations, and extracting the characteristics of each historical wind speed fluctuation;
marking a maximum value and a minimum value in the wind speed waveform corresponding to the historical period, enabling a waveform segment consisting of adjacent minimum values, maximum values and minimum values to be a wind speed fluctuation, and dividing the wind speed waveform corresponding to the historical period into a plurality of historical wind speed fluctuations;
b-2, extracting the characteristics of each historical wind speed fluctuation, wherein the characteristics comprise: minimum value, maximum value, time length of two minimum values in wind speed fluctuation, ratio of slope of wind speed ascending section and wind speed descending section, average value and standard deviation.
Step C: intercepting power fluctuation corresponding to each historical wind speed fluctuation from the power waveform with the highest wind speed waveform similarity corresponding to the historical period;
step D: and uniformly processing the power fluctuation corresponding to each historical wind speed fluctuation into power fluctuation with the same time length by using a cubic spline interpolation method.
Step E: and taking the characteristic value of each historical wind speed fluctuation as input data of the initial BPNN, taking the power fluctuation corresponding to each historical wind speed fluctuation as output data of the initial BPNN, training the initial BPNN, and obtaining a wind power prediction model.
The method comprises the following steps of clustering historical wind speed fluctuation based on the characteristics of the historical wind speed fluctuation, and obtaining clusters and cluster centers of the clusters, wherein the clustering comprises the following steps:
and clustering the historical wind speed fluctuation by adopting a clustering method such as a k-means clustering algorithm and the like based on the characteristics of the historical wind speed fluctuation to obtain clusters and cluster centers of the clusters.
The process for acquiring the fitting coefficients A and B corresponding to each cluster comprises the following steps:
acquiring actual power data and predicted power data at each moment in the historical wind speed fluctuation in the ith cluster, wherein the predicted power data at each moment in the historical wind speed fluctuation in the ith cluster are power data at corresponding moments in power fluctuation corresponding to the historical wind speed fluctuation in the ith cluster;
based on the actual power data and the predicted power data at each moment, a linear least square method is adopted to train a linear regression equation: the actual wind power is Axpredicted wind power + B, and a fitting coefficient A corresponding to the ith cluster is obtained i And B i ；
Wherein i belongs to (1-N), and N is the total number of clusters.
After verification, a comparison schematic diagram as shown in fig. 4 is obtained, in the diagram, compared with the power prediction result adopting the method of the present invention, the power prediction result adopting the traditional method has a larger deviation with the actual power and the predicted power, and the effectiveness of the method of the present invention is illustrated.
The invention provides a wind power short-term power prediction method based on fluctuation identification and error correction, which is based on historical wind speed forecast data, divides the wind power short-term power into a plurality of data sections according to a wind speed fluctuation process and extracts characteristic values; extracting power data at the same time as the wind speed forecast data segment, uniformly interpolating the power data at different time intervals into a sequence with a fixed length, training a correlation model of a wind speed characteristic value and the power sequence, and establishing a prediction model; in the prediction execution, the power prediction model takes the characteristic value of the divided fluctuation process as input, takes the power sequence as output, and obtains a target prediction power value through reconstruction; and establishing correction models of different fluctuation processes through classification and matching of the fluctuation processes, predicting power and correcting to finally obtain a predicted value of the wind power. The wind power prediction is realized by dividing the process of forecasting the wind speed and extracting the characteristics and establishing a correlation model of the wind process and the power, so that the fluctuation of the wind power can be effectively captured, and the prediction level is improved.
According to the method, the wind power prediction result is corrected, and a correction equation is obtained by fitting the basic principle according to the deviation between the predicted power value and the actual power value, so that the correction of the wind power prediction result can be realized by adopting various statistical methods: including but not limited to neural networks, non-linear regression, etc.
The wind power short-term power prediction method based on fluctuation recognition and error correction provided by the invention considers theoretical technology and practical feasibility overall, has a complete technical system, can be widely applied to prediction systems of a wind power plant end and a dispatching end after the achievement is mature through theoretical attack and trial verification, and is calculated according to the scale of the current wind power plant and future development, and the achievement has huge application scale and prospect.
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 has been 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 flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations 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 (11)
1. A wind power short-term power prediction method based on fluctuation identification and error correction is characterized by comprising the following steps:
dividing the wind speed waveform of the duration to be predicted into a plurality of wind speed fluctuations, and forming a characteristic sequence of each wind speed fluctuation;
obtaining low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation by utilizing a pre-established wind power short-term power prediction model and the characteristic sequence of the wind speed fluctuation;
generating wind power fluctuation corresponding to each wind speed fluctuation according to the low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation;
and the wind speed waveform of the duration to be predicted is generated based on wind speed forecast data at each moment in the duration to be predicted.
2. The method of claim 1, wherein the training process of the pre-established wind power short-term power prediction model comprises:
dividing the wind speed waveform of the historical duration into a plurality of wind speed fluctuations, and forming a characteristic sequence of each wind speed fluctuation;
preprocessing the actual wind power waveform with the historical duration to obtain each low-frequency power fluctuation corresponding to each wind speed fluctuation time divided by the wind speed waveform with the historical duration;
uniformly processing the low-frequency power fluctuation into low-frequency power fluctuation with the same time length by using a cubic spline interpolation method;
and as input data of the initial BPNN neural network, taking each low-frequency power fluctuation after interpolation as output data of the initial BPNN neural network, and training the initial BPNN neural network to obtain a wind power short-term power prediction model.
3. The method of claim 1 or 2, wherein the determination of the wind speed fluctuation comprises:
marking a maximum value and a minimum value in the wind speed waveform;
and according to the principle that the adjacent waveform segment corresponding to the minimum value-maximum value-minimum value is a fluctuation, dividing the wind speed waveform into a plurality of wind speed fluctuations.
4. The method of claim 2, wherein the preprocessing the historical-duration actual wind power waveform to obtain each low-frequency power fluctuation corresponding to each wind speed fluctuation time divided by the historical-duration wind speed waveform comprises:
performing wavelet decomposition on the actual wind power waveform with the historical duration by adopting a db6 wavelet decomposition technology to obtain each layer of profile coefficients;
reconstructing the profile coefficients of each layer to obtain low-frequency power waveforms corresponding to the profile coefficients of each layer;
respectively calculating Pearson correlation coefficients between low-frequency power waveforms corresponding to the profile coefficients of each layer and actual wind power waveforms of historical duration, and selecting a low-frequency power waveform with the highest correlation coefficient;
and intercepting the selected low-frequency power waveform to obtain each low-frequency power fluctuation corresponding to each wind speed fluctuation time divided by the wind speed waveform with the historical duration.
5. The method according to claim 1, wherein the generating wind power fluctuation corresponding to each wind speed fluctuation according to the low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation comprises:
carrying out reverse interpolation processing on low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation;
and correcting the power value in each low-frequency power fluctuation after the reverse interpolation processing to obtain the wind power fluctuation corresponding to each wind speed fluctuation.
6. The method as claimed in claim 5, wherein the performing of the inverse interpolation process on the low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation comprises:
and (3) reversely interpolating the low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation into low-frequency power fluctuation consistent with the time length of each wind speed fluctuation by adopting a cubic spline interpolation method, and adjusting the time node of the low-frequency power fluctuation to be consistent with the corresponding wind speed fluctuation.
7. The method of claim 6, wherein the modifying the power value in each low frequency power fluctuation after the reverse interpolation process comprises:
determining a cluster to which the wind speed fluctuation belongs based on a characteristic sequence of the wind speed fluctuation corresponding to the low-frequency power fluctuation after reverse interpolation processing;
calling a first correction coefficient and a second correction coefficient corresponding to the cluster;
substituting the first correction coefficient and the second correction coefficient corresponding to the cluster and the power value in the low-frequency power fluctuation into a correction equation to obtain a correction value of the power value in the low-frequency power fluctuation;
wherein the clustering is based on clustering of individual wind speed fluctuations demarcated from historical wind speed waveforms.
8. The method of claim 7, wherein determining the first correction factor and the second correction factor for the cluster comprises:
substituting the characteristic sequence of the wind speed fluctuation contained in the cluster into a pre-established wind power short-term power prediction model to obtain low-frequency power fluctuation corresponding to the wind speed fluctuation contained in the cluster;
carrying out reverse interpolation processing on low-frequency power fluctuation corresponding to wind speed fluctuation contained in the clusters;
and taking the actual wind power at each moment in the wind speed fluctuation contained in the cluster as a dependent variable in a correction equation, taking the power value at the corresponding moment in the low-frequency power fluctuation corresponding to the wind speed fluctuation contained in the cluster after reverse interpolation processing as an independent variable in the correction equation, and fitting by adopting a linear least square method to obtain a first correction coefficient and a second correction coefficient corresponding to the cluster.
9. The method of claim 7, wherein the correction equation is:
the corrected value of the power value is A multiplied by the power value + B;
wherein A is a first fitting coefficient, and B is a second fitting coefficient.
10. The method of claim 1, wherein the features in the sequence of features of wind speed fluctuations include, but are not limited to:
minimum value, maximum value, time length between two minimum values in wind speed fluctuation, ratio of slope of wind speed ascending section and wind speed descending section, mean value and standard deviation.
11. A wind power short-term power prediction system based on fluctuation identification and error correction, the system comprising:
the dividing module is used for dividing the wind speed waveform of the duration to be predicted into a plurality of wind speed fluctuations and forming a characteristic sequence of each wind speed fluctuation;
the acquisition module is used for acquiring low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation by utilizing a pre-established wind power short-term power prediction model and the characteristic sequence of the wind speed fluctuation;
the generating module is used for generating wind power fluctuation corresponding to each wind speed fluctuation according to the low-frequency power fluctuation corresponding to the characteristic sequence of each wind speed fluctuation;
the wind speed waveform of the time length to be predicted is generated based on wind speed forecast data at each moment in the time length to be predicted.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110240352.4A CN115034422A (en) | 2021-03-04 | 2021-03-04 | Wind power short-term power prediction method and system based on fluctuation identification and error correction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110240352.4A CN115034422A (en) | 2021-03-04 | 2021-03-04 | Wind power short-term power prediction method and system based on fluctuation identification and error correction |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115034422A true CN115034422A (en) | 2022-09-09 |
Family
ID=83117837
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110240352.4A Pending CN115034422A (en) | 2021-03-04 | 2021-03-04 | Wind power short-term power prediction method and system based on fluctuation identification and error correction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115034422A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117829202A (en) * | 2024-03-05 | 2024-04-05 | 西安热工研究院有限公司 | Energy storage auxiliary black start wind speed prediction method and system based on double errors |
| CN117829202B (en) * | 2024-03-05 | 2024-05-10 | 西安热工研究院有限公司 | Energy storage auxiliary black start wind speed prediction method and system based on double errors |
-
2021
- 2021-03-04 CN CN202110240352.4A patent/CN115034422A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117829202A (en) * | 2024-03-05 | 2024-04-05 | 西安热工研究院有限公司 | Energy storage auxiliary black start wind speed prediction method and system based on double errors |
| CN117829202B (en) * | 2024-03-05 | 2024-05-10 | 西安热工研究院有限公司 | Energy storage auxiliary black start wind speed prediction method and system based on double errors |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Huang et al. | Forecasting hourly solar irradiance using hybrid wavelet transformation and Elman model in smart grid | |
| CN102663513B (en) | Utilize the wind power combined prediction modeling method of grey relational grade analysis | |
| CN111080032A (en) | Load prediction method based on Transformer structure | |
| CN105574615B (en) | wavelet-BP neural network wind power prediction method based on spatial correlation and GA | |
| CN107704953A (en) | The short-term wind-electricity power probability density Forecasting Methodology of EWT quantile estimate forests | |
| Dou et al. | Hybrid model for renewable energy and loads prediction based on data mining and variational mode decomposition | |
| CN111027775A (en) | Step hydropower station generating capacity prediction method based on long-term and short-term memory network | |
| CN103020743B (en) | Wind energy turbine set ultra-short term wind speed forecasting method | |
| Yang et al. | A deep attention convolutional recurrent network assisted by k-shape clustering and enhanced memory for short term wind speed predictions | |
| CN112396234A (en) | User side load probability prediction method based on time domain convolutional neural network | |
| CN116307291B (en) | Distributed photovoltaic power generation prediction method and prediction terminal based on wavelet decomposition | |
| CN113822418A (en) | Wind power plant power prediction method, system, device and storage medium | |
| CN114819374A (en) | Regional new energy ultra-short term power prediction method and system | |
| Chen et al. | Research on wind power prediction method based on convolutional neural network and genetic algorithm | |
| CN105956708A (en) | Grey correlation time sequence based short-term wind speed forecasting method | |
| CN115358437A (en) | Power supply load prediction method based on convolutional neural network | |
| CN104933471A (en) | Phase space reconstruction method for use in short-term wind speed prediction method | |
| CN104834816A (en) | Short-term wind speed prediction method | |
| CN117634678A (en) | Low-carbon park carbon emission prediction method based on actual operation scene | |
| CN109063930A (en) | A kind of dynamic wind power plant general power prediction technique based on clustering | |
| CN111786391A (en) | Probabilistic power flow analysis method and system considering energy uncertainty | |
| CN115034422A (en) | Wind power short-term power prediction method and system based on fluctuation identification and error correction | |
| CN109840308B (en) | Regional wind power probability forecasting method and system | |
| CN110852597B (en) | Electricity consumption peak period resident load ratio calculation method based on generation of countermeasure network | |
| CN114372640A (en) | Wind power prediction method based on fluctuation sequence classification correction |
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 |