CN113964825A - Ultrashort-term wind power prediction method based on secondary decomposition and BiGRU - Google Patents
Ultrashort-term wind power prediction method based on secondary decomposition and BiGRU Download PDFInfo
- Publication number
- CN113964825A CN113964825A CN202111228476.7A CN202111228476A CN113964825A CN 113964825 A CN113964825 A CN 113964825A CN 202111228476 A CN202111228476 A CN 202111228476A CN 113964825 A CN113964825 A CN 113964825A
- Authority
- CN
- China
- Prior art keywords
- wind power
- decomposition
- bigru
- time sequence
- power prediction
- 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
- 238000000354 decomposition reaction Methods 0.000 title claims abstract description 91
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000001228 spectrum Methods 0.000 claims abstract description 13
- 230000004927 fusion Effects 0.000 claims abstract description 10
- 238000004590 computer program Methods 0.000 claims description 14
- 238000010606 normalization Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 6
- 238000010248 power generation Methods 0.000 abstract description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000013135 deep learning Methods 0.000 description 4
- 238000007619 statistical method Methods 0.000 description 4
- 238000000053 physical method Methods 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 101100045541 Homo sapiens TBCD gene Proteins 0.000 description 1
- 101150093640 SSD1 gene Proteins 0.000 description 1
- 101100111629 Saccharomyces cerevisiae (strain ATCC 204508 / S288c) KAR2 gene Proteins 0.000 description 1
- 102100030290 Tubulin-specific chaperone D Human genes 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000002156 mixing Methods 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- 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
-
- 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
-
- 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
- H02J2203/10—Power transmission or distribution systems management focussing at grid-level, e.g. load flow analysis, node profile computation, meshed network optimisation, active network management or spinning reserve management
-
- 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
- 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)
- Business, Economics & Management (AREA)
- Economics (AREA)
- Power Engineering (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Wind Motors (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Control Of Eletrric Generators (AREA)
Abstract
The invention relates to an ultra-short-term wind power prediction method based on secondary decomposition and BiGRU. The method is suitable for the field of wind power generation power prediction. The technical scheme adopted by the invention is as follows: an ultrashort-term wind power prediction method based on secondary decomposition and BiGRU is characterized in that: acquiring wind power data, measuring the correlation degree between each variable and wind power in the wind power data by adopting a Pearson correlation coefficient, and selecting the variable to form input information based on the Pearson coefficient; normalizing the input information; decomposing the normalized input information mode by a secondary decomposition method combining singular spectrum decomposition and variational mode decomposition; respectively modeling time sequence information of all mode signals subjected to mode decomposition by adopting a BiGRU; and the outputs of all the time sequence information networks are jointly input to a full-connection network layer for information fusion and decision making, and a final wind power prediction result is obtained.
Description
Technical Field
The invention relates to an ultra-short-term wind power prediction method based on secondary decomposition and BiGRU. The method is suitable for the field of wind power generation power prediction.
Background
Wind energy is a novel energy source, and is widely used due to the characteristics of unlimited reserve, safety, cleanness and the like, so that the wind energy is vigorously developed in various countries. The randomness and the fluctuation of the wind power generation power are high due to the non-stationarity of the wind speed, the challenges are brought to the safe, stable and economic operation of large-scale wind power generation grid connection, and the accuracy rate of wind power generation power prediction needs to be improved.
Short-term and ultra-short-term prediction can provide reliable electric power transient information for electric power scheduling and wind power generation grid connection safety, and therefore wind power prediction research is mainly focused on short-term and ultra-short-term wind power prediction.
Wind power prediction is roughly divided into the following four types of (1) physical methods; (2) a statistical method; (3) a deep learning method; (4) and (3) a mixing method.
The physical method is mainly based on Numerical Weather Prediction (NWP), and uses meteorological information and geographic information such as solar irradiance, wind speed and temperature to predict wind power.
The statistical method carries out the prediction of the wind power by establishing the statistical relationship between the weather and other related information and the wind power, and compared with a physical method, the statistical method is simple in modeling and widely applied in the early stage.
With the rapid development of deep learning and great success in various fields, wind power prediction methods based on deep learning are researched in a large quantity. The time series modeling based on the BiGRU network is proved to be good in wind power prediction, and the prediction effect is better than that of a traditional statistical method and a shallow neural network method.
The hybrid model prediction method is a research hotspot in the field of wind power prediction, and the current hybrid models are mainly divided into two types, namely a weather typing hybrid prediction model based on a clustering algorithm and a hybrid prediction model based on signal decomposition. Hybrid predictive models based on signal decomposition methods ("decomposition-prediction-reconstruction" methods) are of increasing interest in wind power prediction. The method can effectively extract the significant characteristics of wind power by wavelet transformation, variational modal decomposition, empirical modal decomposition, singular spectrum analysis and other signal decomposition methods, and is applied to wind power prediction.
Although the prediction method of the wind power is more and more focused by researchers, the existing method still has the problem of insufficient prediction accuracy.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: aiming at the existing problems, the ultrashort-term wind power prediction method based on the quadratic decomposition and the BiGRU is provided.
The technical scheme adopted by the invention is as follows: an ultrashort-term wind power prediction method based on secondary decomposition and BiGRU is characterized in that:
acquiring wind power data, measuring the correlation degree between each variable and wind power in the wind power data by adopting a Pearson correlation coefficient, and selecting the variable to form input information based on the Pearson coefficient;
normalizing the input information;
decomposing the normalized input information mode by a secondary decomposition method combining singular spectrum decomposition and variational mode decomposition;
respectively modeling time sequence information of all mode signals subjected to mode decomposition by adopting a BiGRU;
and the outputs of all the time sequence information networks are jointly input to a full-connection network layer for information fusion and decision making, and a final wind power prediction result is obtained.
The input information comprises warp wind speed, weft wind speed and historical power.
The secondary decomposition method combining singular spectrum decomposition and variational mode decomposition is adopted to carry out mode decomposition on the normalized input information, and comprises the following steps:
and after singular spectrum decomposition, the normalized input information reserves the main signal and the high-frequency signal, and the high-frequency signal is decomposed into k mode signals by utilizing variational mode decomposition.
The modeling of the time sequence information of all the mode signals after the mode decomposition by adopting the BiGRU respectively comprises the following steps:
and respectively carrying out time sequence information modeling on the main signal and the k mode signals by adopting the BiGRU.
The output of all time sequence information networks is input to a full-connection network layer together for information fusion and decision, and the method comprises the following steps:
and the outputs of all the time sequence information networks are input into two fully-connected network layers together, the number of nodes of the first fully-connected layer corresponds to the number of decomposed mode signals, the number of nodes of the second fully-connected layer is 32, and finally, a wind power prediction result is output.
The time sequence information modeling comprises the selection of the length of a time sequence signal, the length of the time sequence signal depends on the degree of correlation between the input information of historical moments and future wind power, the degree of correlation between the input information of the historical moments and the future wind power is represented by a Pearson coefficient, and the length of the time sequence of a time sequence is selected based on the Pearson coefficient.
An ultra-short-term wind power prediction device based on quadratic decomposition and BiGRU is characterized in that:
the input information acquisition module is used for acquiring wind power data, measuring the correlation degree between each variable in the wind power data and wind power by adopting a Pearson correlation coefficient, and selecting the variable to form input information based on the Pearson coefficient;
the normalization processing module is used for normalizing the input information;
the second decomposition is used for decomposing the normalized input information mode by adopting a second decomposition method combining singular spectrum decomposition and variational mode decomposition;
the modeling module is used for respectively modeling the time sequence information of all the mode signals after the mode decomposition by adopting the BiGRU;
and the result prediction module is used for inputting the outputs of all the time sequence information networks to the full-connection network layer together for information fusion and decision making to obtain a final wind power prediction result.
A storage medium having stored thereon a computer program executable by a processor, the computer program comprising: the computer program, when executed, implements the steps of the quadratic decomposition and BiGRU based ultra-short term wind power prediction method.
An ultra-short-term wind power prediction device having a memory and a processor, the memory having stored thereon a computer program executable by the processor, characterized by: the computer program, when executed, implements the steps of the quadratic decomposition and BiGRU based ultra-short term wind power prediction method.
The invention has the beneficial effects that: according to the method, variables are selected from wind power data to form input information based on Pearson coefficients, a secondary decomposition method combining singular spectrum decomposition and variational mode decomposition is adopted to carry out mode decomposition on the input information, k mode signals generated by decomposition of main signals and high-frequency signals are reserved, BiGRU is adopted to carry out time sequence information modeling on all mode signals after mode decomposition, the outputs of all time sequence information networks are jointly input to a full-connection network layer to carry out information fusion and decision making, a final wind power prediction result is obtained, and the accuracy rate of wind power prediction is effectively improved.
The decomposed main signal reflects the internal change trend of the original signal, the high-frequency signal reflects the random fluctuation condition, the high-frequency signal is further decomposed, the fluctuation of the signal is effectively reduced, and the complexity of wind power time sequence feature mapping is reduced.
Drawings
FIG. 1 is a flow chart of an ultra-short-term wind power prediction algorithm in the embodiment.
FIG. 2 is a diagram illustrating the SSD decomposition results of the original signal in the example.
FIG. 3 is a diagram showing the result of VMD decomposition of the high frequency signal in the example.
Fig. 4 is an internal structural view of the GRU unit.
Fig. 5 is a structural diagram of a BiGRU network.
Detailed Description
As shown in fig. 1, the present embodiment is an ultra-short-term wind power prediction method based on quadratic decomposition and BiGRU, which specifically includes the following steps:
s1, wind power data are obtained, correlation degrees between variables in the wind power data and wind power are measured through Pearson correlation coefficients, the variables are selected to form input information based on the Pearson coefficients, and the Pearson coefficients are larger than 0.6.
The wind power plant can acquire various wind power data such as environment temperature, air pressure, warp-wise wind speed, weft-wise wind speed, relative humidity, actual wind power and the like at each data acquisition moment. In order to extract information related to wind power prediction and eliminate interference of useless information, in this embodiment, a Pearson correlation coefficient is used to measure a correlation degree between each variable and wind power, and a calculation formula of the Pearson correlation coefficient is as follows:
in the formula
Is the average of the variables x, y.
Through calculation, Pearson coefficient values between the warp-wise wind speed, the weft-wise wind speed and the wind power are respectively 0.873 and 0.895, and strong correlation is achieved. Since the historical wind power has a time correlation with the future wind power, the radial wind speed, the latitudinal wind speed and the historical wind power are selected as input information in the present example.
And S2, normalizing the input information. In order to eliminate the dimensional influence between the data, a data Normalization process is required, and in this embodiment, Min-Max Normalization (Min-Max Normalization) Normalization is performed on all the variables, as shown in the following formula:
and S3, performing mode decomposition on the normalized input information by adopting a secondary decomposition method combining singular spectrum decomposition and variational mode decomposition.
In order to cope with the influence of the randomness and the volatility of the wind power data on the prediction effect, the present embodiment performs data signal processing by using a secondary decomposition method combining Singular Spectrum Decomposition (SSD) and Variational Mode Decomposition (VMD).
The original signal is subjected to SSD decomposition, and the result is shown in fig. 2, where SSD1 represents the decomposed main signal, SSD2 is the decomposed high frequency signal, SSD3 is noise, and the present embodiment only retains the main signal and the high frequency signal. It can be seen in fig. 2 that the main signal reflects the intrinsic trend of the original signal, and the high frequency signal reflects the random fluctuation.
As shown in fig. 3, VMD decomposition is performed on the high-frequency signal to obtain K (in this example, K takes 5) signals with different center frequencies, and further decomposition of the high-frequency signal can effectively retain main information and eliminate interference information, thereby effectively reducing the volatility of the signal and reducing the complexity of wind power time series feature mapping.
And S4, respectively carrying out time sequence information modeling on all mode signals after mode decomposition by adopting the BiGRU to form a plurality of time sequence information networks corresponding to all mode signals one to one.
The time sequence information modeling is a mainstream solution for processing a time sequence problem, and in the embodiment, a gate control logic unit GRU in deep learning is adopted to construct a bidirectional time sequence information network (BiGRU), and all model signals obtained by secondary decomposition are subjected to time sequence modeling. The selection of the length of the time sequence signal depends on the degree of correlation between the input information at the historical time and the future wind power, the Pearson coefficient is adopted in the embodiment to represent the degree of correlation between the input information at the historical time and the future wind power, the time sequence length of the time sequence is selected based on the Pearson coefficient, and the Pearson coefficient is greater than 0.6.
As shown in FIG. 4, which is an internal structure diagram of GRU, xtAnd ytInput and output at time t, ht-1And htThe hidden states at time t-1 and time t, respectively, σ represents the sigmoid function, and the reset gate and the update gate determine the degree to which previous information is discarded and which information is updated, respectively.
Zt＝σ(Wxzxt+Whzht-1+bz)
rt＝σ(Wxrxt+Whrht-1+br)
The BiGRU network is formed by two opposite-direction GRU networks, and fig. 5 is a BiGRU network structure diagram formed by n GRU units. X ═ X1,x2,…,xn]Is an input timing signal of length n, Y ═ Y1,y2,…,yn]Is the corresponding output result. As can be seen from the figure, compared with the GRU network, the node of the BiGRU network at each time contains the information of the whole input sequence, and can better perform feature expression on the whole input sequence.
And S5, inputting the outputs of all the time sequence information networks into the two fully-connected network layers together for information fusion and decision making, and obtaining a final wind power prediction result. The number of the nodes of the first full-connection layer is K +1 (namely the number of the mode signals after secondary decomposition), the number of the nodes of the second full-connection layer is 32, and finally, a wind power prediction result is output.
The embodiment is an ultra-short-term wind power prediction device based on secondary decomposition and BiGRU, and comprises an input information acquisition module, a normalization processing module, a secondary decomposition and modeling module and a result prediction module.
In the embodiment, the input information acquisition module is used for acquiring wind power data, measuring the correlation degree between each variable in the wind power data and wind power by using a Pearson correlation coefficient, and selecting the variable to form input information based on the Pearson coefficient; the normalization processing module is used for normalizing the input information; the secondary decomposition is used for decomposing the normalized input information mode by adopting a secondary decomposition method combining singular spectrum decomposition and variational mode decomposition; the modeling module is used for respectively modeling the time sequence information of all the mode signals after the mode decomposition by adopting the BiGRU; and the result prediction module is used for inputting the outputs of all the time sequence information networks to the full-connection network layer together for information fusion and decision making to obtain a final wind power prediction result.
The present embodiment also provides a storage medium having stored thereon a computer program executable by a processor, the computer program when executed implementing the steps of the quadratic decomposition and BiGRU based ultra-short term wind power prediction method in this example.
The embodiment also provides an ultra-short-term wind power prediction device, which comprises a memory and a processor, wherein the memory stores a computer program capable of being executed by the processor, and the computer program realizes the steps of the ultra-short-term wind power prediction method based on quadratic decomposition and BiGRU in the embodiment when being executed.
Claims (9)
1. An ultrashort-term wind power prediction method based on secondary decomposition and BiGRU is characterized in that:
acquiring wind power data, measuring the correlation degree between each variable and wind power in the wind power data by adopting a Pearson correlation coefficient, and selecting the variable to form input information based on the Pearson coefficient;
normalizing the input information;
decomposing the normalized input information mode by a secondary decomposition method combining singular spectrum decomposition and variational mode decomposition;
respectively modeling time sequence information of all mode signals subjected to mode decomposition by adopting a BiGRU;
and the outputs of all the time sequence information networks are jointly input to a full-connection network layer for information fusion and decision making, and a final wind power prediction result is obtained.
2. The ultra-short-term wind power prediction method based on quadratic decomposition and BiGRU according to claim 1, characterized in that: the input information comprises warp wind speed, weft wind speed and historical power.
3. The ultra-short-term wind power prediction method based on quadratic decomposition and BiGRU according to claim 1, wherein the quadratic decomposition method combining singular spectrum decomposition and variational mode decomposition is used for carrying out mode decomposition on the normalized input information, and comprises the following steps:
and after singular spectrum decomposition, the normalized input information reserves the main signal and the high-frequency signal, and the high-frequency signal is decomposed into k mode signals by utilizing variational mode decomposition.
4. The ultra-short-term wind power prediction method based on quadratic decomposition and BiGRU of claim 3, wherein the modeling of the time sequence information of all mode signals after the mode decomposition by using the BiGRU respectively comprises:
and respectively carrying out time sequence information modeling on the main signal and the k mode signals by adopting the BiGRU.
5. The ultra-short-term wind power prediction method based on quadratic decomposition and BiGRU according to claim 1, 3 or 4, wherein the output of all timing information networks is commonly input to a fully-connected network layer for information fusion and decision, comprising:
and the outputs of all the time sequence information networks are input into two fully-connected network layers together, the number of nodes of the first fully-connected layer corresponds to the number of decomposed mode signals, the number of nodes of the second fully-connected layer is 32, and finally, a wind power prediction result is output.
6. The ultra-short-term wind power prediction method based on quadratic decomposition and BiGRU according to claim 1, characterized in that: the time sequence information modeling comprises the selection of the length of a time sequence signal, the length of the time sequence signal depends on the degree of correlation between the input information of historical moments and future wind power, the degree of correlation between the input information of the historical moments and the future wind power is represented by a Pearson coefficient, and the length of the time sequence of a time sequence is selected based on the Pearson coefficient.
7. An ultra-short-term wind power prediction device based on quadratic decomposition and BiGRU is characterized in that:
the input information acquisition module is used for acquiring wind power data, measuring the correlation degree between each variable in the wind power data and wind power by adopting a Pearson correlation coefficient, and selecting the variable to form input information based on the Pearson coefficient;
the normalization processing module is used for normalizing the input information;
the second decomposition is used for decomposing the normalized input information mode by adopting a second decomposition method combining singular spectrum decomposition and variational mode decomposition;
the modeling module is used for respectively modeling the time sequence information of all the mode signals after the mode decomposition by adopting the BiGRU;
and the result prediction module is used for inputting the outputs of all the time sequence information networks to the full-connection network layer together for information fusion and decision making to obtain a final wind power prediction result.
8. A storage medium having stored thereon a computer program executable by a processor, the computer program comprising: the computer program when executed implements the steps of the quadratic decomposition and BiGRU based ultra-short term wind power prediction method of any of claims 1 to 6.
9. An ultra-short-term wind power prediction device having a memory and a processor, the memory having stored thereon a computer program executable by the processor, characterized by: the computer program when executed implements the steps of the quadratic decomposition and BiGRU based ultra-short term wind power prediction method of any of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111228476.7A CN113964825A (en) | 2021-10-21 | 2021-10-21 | Ultrashort-term wind power prediction method based on secondary decomposition and BiGRU |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111228476.7A CN113964825A (en) | 2021-10-21 | 2021-10-21 | Ultrashort-term wind power prediction method based on secondary decomposition and BiGRU |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113964825A true CN113964825A (en) | 2022-01-21 |
Family
ID=79465462
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111228476.7A Pending CN113964825A (en) | 2021-10-21 | 2021-10-21 | Ultrashort-term wind power prediction method based on secondary decomposition and BiGRU |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113964825A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113496102A (en) * | 2021-07-05 | 2021-10-12 | 华北电力大学（保定） | Distribution network ultra-short term power situation prediction method based on improved BiGRU |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150302313A1 (en) * | 2014-04-22 | 2015-10-22 | State Grid Corporation Of China | Method of predicating ultra-short-term wind power based on self-learning composite data source |
| CN109214575A (en) * | 2018-09-12 | 2019-01-15 | 河海大学 | A kind of super short-period wind power prediction technique based on small wavelength short-term memory network |
| CN112329979A (en) * | 2020-09-23 | 2021-02-05 | 燕山大学 | Ultra-short-term wind power prediction method based on self-adaptive depth residual error network |
| CN112733462A (en) * | 2021-01-21 | 2021-04-30 | 国网辽宁省电力有限公司阜新供电公司 | Ultra-short-term wind power plant power prediction method combining meteorological factors |
| CN112836434A (en) * | 2021-02-22 | 2021-05-25 | 中国电建集团华东勘测设计研究院有限公司 | Wind power ultra-short-term power prediction method integrating time sequence characteristics and statistical characteristics |
| CN112990553A (en) * | 2021-02-23 | 2021-06-18 | 中国电建集团华东勘测设计研究院有限公司 | Wind power ultra-short-term power prediction method using self-attention mechanism and bilinear fusion |
-
2021
- 2021-10-21 CN CN202111228476.7A patent/CN113964825A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150302313A1 (en) * | 2014-04-22 | 2015-10-22 | State Grid Corporation Of China | Method of predicating ultra-short-term wind power based on self-learning composite data source |
| CN109214575A (en) * | 2018-09-12 | 2019-01-15 | 河海大学 | A kind of super short-period wind power prediction technique based on small wavelength short-term memory network |
| CN112329979A (en) * | 2020-09-23 | 2021-02-05 | 燕山大学 | Ultra-short-term wind power prediction method based on self-adaptive depth residual error network |
| CN112733462A (en) * | 2021-01-21 | 2021-04-30 | 国网辽宁省电力有限公司阜新供电公司 | Ultra-short-term wind power plant power prediction method combining meteorological factors |
| CN112836434A (en) * | 2021-02-22 | 2021-05-25 | 中国电建集团华东勘测设计研究院有限公司 | Wind power ultra-short-term power prediction method integrating time sequence characteristics and statistical characteristics |
| CN112990553A (en) * | 2021-02-23 | 2021-06-18 | 中国电建集团华东勘测设计研究院有限公司 | Wind power ultra-short-term power prediction method using self-attention mechanism and bilinear fusion |
Non-Patent Citations (1)
| Title |
|---|
| LING XIANG等: "Deterministic and probabilistic multi-step forecasting for short-term wind speed based on secondary decomposition and a deep learning method", 《ENERGY CONVERSION AND MANAGEMENT》, 20 June 2020 (2020-06-20), pages 1 - 12 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113496102A (en) * | 2021-07-05 | 2021-10-12 | 华北电力大学（保定） | Distribution network ultra-short term power situation prediction method based on improved BiGRU |
| CN113496102B (en) * | 2021-07-05 | 2024-07-09 | 华北电力大学 | Distribution network ultra-short-term power situation prediction method based on improvement BiGRU |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhang et al. | Short-term forecasting and uncertainty analysis of wind turbine power based on long short-term memory network and Gaussian mixture model | |
| Deng et al. | Wind power forecasting methods based on deep learning: A survey | |
| Wang et al. | A hybrid forecasting model based on outlier detection and fuzzy time series–A case study on Hainan wind farm of China | |
| CN107730044A (en) | A kind of hybrid forecasting method of renewable energy power generation and load | |
| Yang et al. | A deep attention convolutional recurrent network assisted by k-shape clustering and enhanced memory for short term wind speed predictions | |
| CN112100911B (en) | Solar radiation prediction method based on depth BILSTM | |
| CN114970952B (en) | Photovoltaic output short-term prediction method and system considering environmental factors | |
| CN116937579A (en) | Wind power interval prediction considering space-time correlation and interpretable method thereof | |
| Yu et al. | Application of small sample virtual expansion and spherical mapping model in wind turbine fault diagnosis | |
| CN114021483A (en) | Ultra-short-term wind power prediction method based on time domain characteristics and XGboost | |
| CN115995810A (en) | Wind power prediction method and system considering weather fluctuation self-adaptive matching | |
| CN116632820A (en) | Wind power prediction method based on sequence extension VMD and wind speed fluctuation characteristics | |
| CN113964825A (en) | Ultrashort-term wind power prediction method based on secondary decomposition and BiGRU | |
| Xu et al. | Short-term wind speed prediction based on GRU | |
| Cao et al. | Deep learning-based online small signal stability assessment of power systems with renewable generation | |
| CN112418504B (en) | Wind speed prediction method based on mixed variable selection optimization deep belief network | |
| CN117688362A (en) | Photovoltaic power interval prediction method and device based on multivariate data feature enhancement | |
| CN113988395A (en) | Wind power ultra-short-term power prediction method based on SSD and dual attention mechanism BiGRU | |
| Zhao et al. | Selective sampling using active learning for short-term wind speed prediction | |
| Karamdel et al. | A regression model-based short-term PV power generation forecasting | |
| Liao et al. | Wind power prediction based on periodic characteristic decomposition and multi-layer attention network | |
| CN116070768A (en) | Short-term wind power prediction method based on data reconstruction and TCN-BiLSTM | |
| Lin et al. | A Novel Multi-Model Stacking Ensemble Learning Method for Metro Traction Energy Prediction | |
| CN113537573B (en) | Wind power operation trend prediction method based on double space-time feature extraction | |
| CN113449466A (en) | Solar radiation prediction method and system for optimizing RELM based on PCA and chaos GWO |
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 |