CN116488158B - Method and device for predicting medium-long-term electric quantity of wind power based on transfer learning - Google Patents
Method and device for predicting medium-long-term electric quantity of wind power based on transfer learning Download PDFInfo
- Publication number
- CN116488158B CN116488158B CN202310446257.9A CN202310446257A CN116488158B CN 116488158 B CN116488158 B CN 116488158B CN 202310446257 A CN202310446257 A CN 202310446257A CN 116488158 B CN116488158 B CN 116488158B
- Authority
- CN
- China
- Prior art keywords
- wind power
- matrix
- electric quantity
- quantity prediction
- training
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000013526 transfer learning Methods 0.000 title claims abstract description 22
- 238000000034 method Methods 0.000 title claims abstract description 17
- 238000012549 training Methods 0.000 claims abstract description 44
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 34
- 238000013507 mapping Methods 0.000 claims abstract description 28
- 238000010248 power generation Methods 0.000 claims abstract description 13
- 238000013508 migration Methods 0.000 claims abstract description 10
- 230000005012 migration Effects 0.000 claims abstract description 10
- 230000004927 fusion Effects 0.000 claims abstract description 9
- 238000007781 pre-processing Methods 0.000 claims abstract description 9
- 230000005611 electricity Effects 0.000 claims abstract description 8
- 239000011159 matrix material Substances 0.000 claims description 41
- 230000006870 function Effects 0.000 claims description 7
- 238000005457 optimization Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 3
- 230000007774 longterm Effects 0.000 claims description 2
- 230000011218 segmentation Effects 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 5
- 238000007619 statistical method Methods 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000012300 Sequence Analysis Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000000053 physical method Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- 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
- G06N3/08—Learning methods
- G06N3/096—Transfer learning
-
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas 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
- H02J3/004—Generation forecast, e.g. methods or systems for forecasting future energy generation
-
- 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
- H02J2203/20—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
- 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
Abstract
The invention provides a method and a device for predicting the medium-long-term electric quantity of wind power based on transfer learning, comprising the following steps: dividing historical resource data of an area where a wind power station is located into a plurality of training sets according to years; carrying out data preprocessing on historical power generation data of a wind power station, and then forming a sample set; respectively performing migration learning on each training set and the sample set, and establishing mapping; adding a parameter related to time into the transfer learning algorithm, converting the training set and the sample set according to the mapping corresponding to each training set, and constructing and training an electric quantity prediction model corresponding to the training set by using the converted data; and calculating the weights of all the electric quantity prediction models according to the PSO algorithm to obtain a fused final electric quantity prediction model. According to the method, the transfer learning algorithm is creatively used in wind power medium-long-term electricity quantity prediction, and a good wind power medium-long-term electricity quantity prediction effect is achieved through a model fusion algorithm.
Description
Technical Field
The invention belongs to the technical field of new energy power, and particularly relates to a method and a device for predicting the medium-long-term electric quantity of wind power based on transfer learning.
Background
Different from the prediction of the output electric quantity of a representative year (average years) in the resource evaluation, the prediction of the middle-long-term electric quantity of wind power is realized by a statistical method such as physical modeling or sequence analysis and the like, so that the prediction of the electric energy generation capacity of a wind power plant in a longer time scale such as month, year and the like in the future is realized. The wind power medium-long-term electric quantity prediction can be classified into medium-term prediction and long-term prediction on a time scale. The medium-term prediction refers to prediction with a time scale of week or month, and is mainly used for medium-term power generation scheduling and rolling adjustment of maintenance plans so as to avoid relevant maintenance in a large wind period. The wind power medium-long-term electric quantity prediction is generally classified into a physical method and a statistical method 2 from the viewpoint of model construction. The statistical method further comprises sequence derivation, gray scale model, intelligent learning algorithm and the like.
For the statistical method, whether classical machine learning or current popular deep learning is adopted, the requirement of wind power medium-long-term electric quantity prediction on the quantity of training sets is relatively high (although a linear model in classical machine learning has relatively low requirement on training data compared with other models, the linear model has relatively poor expression capability, data with relatively large distribution difference between training data and test data cannot be well modeled, and the method is not suitable for wind power medium-long-term electric quantity prediction), and in an actual scene, many stations, particularly newer stations, have relatively less historical power generation data, so that many actual scenes cannot meet the requirement on the training data, the algorithm effect is relatively poor, and the method is particularly suitable for deep learning.
Disclosure of Invention
The invention provides a method and a device for predicting the medium-long-term electric quantity of wind power based on transfer learning, which solve the problem of less historical power generation data through transfer learning and a model fusion algorithm, and achieve a higher-precision prediction effect.
In order to achieve the above purpose, the technical scheme of the invention is realized as follows:
a method for predicting the medium-long term electric quantity of wind power based on transfer learning, comprising the following steps:
s1, dividing historical resource data of an area where a wind power station is located into a plurality of training sets according to years; the historical resource data are meteorological data;
s2, carrying out data preprocessing on historical power generation data of the wind power station, and then forming a sample set; the data preprocessing is to process abnormal values in historical power generation data;
s3, respectively performing migration learning on each training set and the sample set, and establishing mapping in another space; adding a parameter related to time into the algorithm of transfer learning, wherein the parameter is set for the dividing year of the historical resource data and is inversely proportional to the time;
s4, converting the training set and the sample set according to the mapping corresponding to each training set, and constructing and training an electric quantity prediction model corresponding to the training set by using the converted data;
and S5, calculating the weight of all the electric quantity prediction models according to a PSO algorithm, and obtaining a fused final electric quantity prediction model.
Further, in step S3, the migration algorithm includes:
let ini be the i-th training set, i=1, 2, … …, N; d is the sample set;
wherein ,is a unitary matrix->Is a central matrix->Is a column matrix, l T A transposed matrix of l; k is a kernel matrix,>wherein each element is RBF kernel, n i For Traini, n is the sample size of D,
l is a parameter matrix, x in the expression of L i ,x j Is a matrix L ij Matrix elements of (a);
W i mapping for Traini, W i T Is W i Is a transposed matrix of (a);
tr () is a trace function representing the sum of elements on the main diagonal of the matrix;
u (t) is a parameter related to the time, and is inversely proportional to the time, and is set for the divided year of the resource data;
according to the algorithm, the mapping W for each Traini is solved by convex optimization i 。
Further, the step S5 specifically includes:
s501, calculating alpha according to PSO algorithm i Such that:
wherein Modeli is the ith wind power electric quantity prediction model, alpha i The weight of the ith wind power electricity quantity prediction model is calculated;
s502, obtaining a wind power medium-long term electric quantity prediction model M,
the invention also provides a device for predicting the medium-long-term electric quantity of wind power based on transfer learning, which comprises:
the segmentation module is used for segmenting historical resource data of the area where the wind power station is located into a plurality of training sets according to years;
the sample set module is used for carrying out data preprocessing on historical power generation data of the wind power station and then forming a sample set;
the mapping module is used for respectively carrying out migration learning on each training set and the sample set and establishing mapping in another space; adding a parameter related to time into the algorithm of transfer learning, wherein the parameter is set for the dividing year of the historical resource data and is inversely proportional to the time;
the model construction module is used for converting the training set and the sample set according to the mapping corresponding to each training set, and constructing and training an electric quantity prediction model corresponding to the training set by using the converted data;
and the fusion module calculates the weight of all the electric quantity prediction models according to the PSO algorithm to obtain a fused final electric quantity prediction model.
Further, the mapping module includes a migration learning unit, where the migration learning unit includes:
let ini be the i-th training set, i=1, 2, … …, N; d is the sample set;
wherein ,is a unitary matrix->Is a central matrix->Is a column matrix, l T A transposed matrix of l; k is a kernel matrix,>wherein each element is RBF kernel, n i For Traini, n is the sample size of D,
l is a parameter matrix, x in the expression of L i ,x j Is a matrix L ij Matrix elements of (a);
W i for the mapping of the trail(s),is W i Is a transposed matrix of (a);
tr () is a trace function representing the sum of elements on the main diagonal of the matrix;
u (t) is a parameter related to the time, and is inversely proportional to the time, and is set for the divided year of the resource data;
according to the algorithm, the mapping W for each Traini is solved by convex optimization i 。
Further, the fusion module includes:
weight unit for calculating alpha according to PSO algorithm i So that
Wherein Modeli is the ith wind power electric quantity prediction model, alpha i The weight of the ith wind power electricity quantity prediction model is calculated;
a fusion unit for obtaining a wind power medium-long term electric quantity prediction model M,
compared with the prior art, the invention has the following beneficial effects:
1. the invention innovatively uses the transfer learning algorithm in wind power medium-long-term electric quantity prediction, provides a way for solving the problem of less historical power generation data by improving the transfer learning algorithm.
2. According to the invention, the time parameter is added in the transfer learning algorithm, so that the prediction precision can be improved, and the algorithm shows a higher precision effect on unknown electric quantity prediction data.
3. The method achieves a good wind power medium-long term electricity quantity prediction effect through a model fusion algorithm.
Drawings
FIG. 1 is a schematic flow chart of an embodiment of the invention.
Detailed Description
It should be noted that, without conflict, the embodiments of the present invention and features of the embodiments may be combined with each other.
The invention will be further described with reference to the drawings and the specific examples.
In this embodiment, a certain wind farm station belongs to a wind farm station with a relatively new and relatively short setup time, and the steps for implementing the medium-long term electric quantity prediction according to the method of the present invention are shown in fig. 1, and include:
1: firstly, dividing 30 years of historical resource data of an area where a wind farm station is located into 30 training sets according to time and year, wherein the training sets are Train1, train2, … and Train30 respectively;
the resource data described in this step is derived from ERA5 (fifth generation atmospheric analysis dataset), and each of the segmented training sets includes: a monthly average wind speed, a monthly average humidity, a monthly average temperature, a monthly average pressure, a monthly average wind direction, etc.
2: secondly, data preprocessing is carried out on the historical power generation data of the station, wherein the preprocessing mainly comprises the steps of eliminating abnormal values in the historical power generation data: including null values, dead values, limit values, etc.
After the abnormal values are removed, the invalid generated energy data can be complemented by an interpolation replacement method.
After data preprocessing, the wind power station has short history and little generated energy data, so that a sample set D is formed by using the preprocessed generated energy data.
3: mapping in another space is established for each training set training (i=1, 2, …, 30) with the sample set D, respectively, according to the following algorithm:
wherein ,is a unitary matrix->Is a central matrix->Is a column matrix, l T A transposed matrix of l; k is a kernel matrix,>wherein each element is RBF kernel, n i For Traini, n is the sample size of D, L is the parameter matrix, and the expression is as follows:
x in the expression of L i ,x j Is a matrix L ij Matrix elements of (a);
W i for the mapping of the trail(s),is W i Is a transposed matrix of (a);
tr () is a trace function representing the sum of elements on the main diagonal of the matrix;
u (t) is a parameter related to time and inversely proportional to time, and is set for the resource data by the year, for example, u (t) is a relatively simple setting method, because the split year selected in this embodiment is 30 years, u (t) of the year farthest from the current time (30 years) is set to 1/30, u (t) of the next year (29 years from now) is 1/29, and so on, the more recent year the current time is, u (t) is larger;
in addition to the above setting method, u (t) may be set by using another function as long as the calculation result of the function is inversely proportional to time according to the division year.
The time parameter u (t) is added in the algorithm, so that the prediction accuracy can be improved.
According to the algorithm, the mapping W for each Traini is solved by convex optimization i 。
4: according to the mapping W i Each set of data sets { Traini, D } comprising a training set of resource data and a historical power generation data sample set D of sites is converted to { TTi, di }, where TTi is Traini according to a mapping W i Converted data, di is D according to the mapping W i Converted data; the sequence number is denoted by i as in the case of Traini, because of the one-to-one conversion.
Modeling after conversion, each set of data { TTi, di } being used to construct and train a model of the electric quantity prediction of one of the wind farm stations, for a total of 30 sets of model i (i=1, 2, …, 30); the electric quantity prediction model can adopt a linear regression model or a gbm tree model or other common prediction models;
5: alpha is calculated according to PSO algorithm i So that
Wherein Modeli is the ith wind power electric quantity prediction model, alpha i The weight of the ith wind power electricity quantity prediction model is calculated;
obtaining a wind power medium-long term electric quantity prediction model M,
the foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (4)
1. A method for predicting the medium-long term electric quantity of wind power based on transfer learning, which is characterized by comprising the following steps:
s1, dividing historical resource data of an area where a wind power station is located into a plurality of training sets according to years;
s2, carrying out data preprocessing on historical power generation data of the wind power station, and then forming a sample set;
s3, respectively performing migration learning on each training set and the sample set, and establishing mapping in another space; adding a parameter related to time into the algorithm of transfer learning, wherein the parameter is set for the dividing year of the historical resource data and is inversely proportional to the time;
s4, converting the training set and the sample set according to the mapping corresponding to each training set, and constructing and training an electric quantity prediction model corresponding to the training set by using the converted data;
s5, calculating weights of all electric quantity prediction models according to a PSO algorithm to obtain a fused final electric quantity prediction model;
in step S3, the algorithm of the transfer learning includes:
let ini be the i-th training set, i=1, 2, … …, N; d is the sample set;
；
wherein ,is a unitary matrix->Is a central matrix of which the number of the pixels is equal,in the form of a column matrix,l T is thatlIs a transposed matrix of (a); k is kernelMatrix (S)>Wherein each element is RBF kernel,
n i for Traini, n is the sample size of D,
；
l is a parameter matrix, x in the expression of L i ,x j Is a matrix L ij Matrix elements of (a);
W i for the mapping of the trail(s),is thatW i Is a transposed matrix of (a);
tr () is a trace function representing the sum of elements on the main diagonal of the matrix;
u (t) is a parameter related to the time, and is inversely proportional to the time, and is set for the divided year of the resource data;
according to the algorithm, the mapping for each Traini is solved by convex optimizationW i 。
2. The method for predicting the mid-long term power of wind power based on transfer learning according to claim 1, wherein step S5 specifically comprises:
s501, calculating alpha according to PSO algorithm i So that
；
Wherein Modeli is the ith wind power electric quantity prediction model, alpha i The weight of the ith wind power electricity quantity prediction model is calculated;
s502, obtaining a wind power medium-long term electric quantity prediction model M,。
3. a device for predicting the medium-long-term electric quantity of wind power based on transfer learning, which is characterized by comprising:
the segmentation module is used for segmenting historical resource data of the area where the wind power station is located into a plurality of training sets according to years;
the sample set module is used for carrying out data preprocessing on historical power generation data of the wind power station and then forming a sample set;
the mapping module is used for respectively carrying out migration learning on each training set and the sample set and establishing mapping in another space; adding a parameter related to time into the algorithm of transfer learning, wherein the parameter is set for the dividing year of the historical resource data and is inversely proportional to the time;
the model construction module is used for converting the training set and the sample set according to the mapping corresponding to each training set, and constructing and training an electric quantity prediction model corresponding to the training set by using the converted data;
the fusion module calculates the weight of all the electric quantity prediction models according to a PSO algorithm to obtain a fused final electric quantity prediction model;
the mapping module comprises a migration learning unit, wherein the migration learning unit comprises:
let ini be the i-th training set, i=1, 2, … …, N; d is the sample set;
；
wherein ,is a unitary matrix->Is a central matrix of which the number of the pixels is equal,in the form of a column matrix,l T is thatlIs a transposed matrix of (a); k is a kernel matrix,>wherein each element is RBF kernel,
n i for Traini, n is the sample size of D,
；
l is a parameter matrix, x in the expression of L i ,x j Is a matrix L ij Matrix elements of (a);
W i for the mapping of the trail(s),is thatW i Is a transposed matrix of (a);
tr () is a trace function representing the sum of elements on the main diagonal of the matrix;
u (t) is a parameter related to the time, and is inversely proportional to the time, and is set for the divided year of the resource data;
according to the algorithm, the mapping for each Traini is solved by convex optimizationW i 。
4. The apparatus for long-term power prediction in wind power based on transfer learning according to claim 3, wherein the fusion module comprises:
weight unit for calculating alpha according to PSO algorithm i So that
；
Wherein Modeli is the ith wind power electric quantity prediction model, alpha i The weight of the ith wind power electricity quantity prediction model is calculated;
a fusion unit for obtaining a wind power medium-long term electric quantity prediction model M,。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310446257.9A CN116488158B (en) | 2023-04-24 | 2023-04-24 | Method and device for predicting medium-long-term electric quantity of wind power based on transfer learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310446257.9A CN116488158B (en) | 2023-04-24 | 2023-04-24 | Method and device for predicting medium-long-term electric quantity of wind power based on transfer learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116488158A CN116488158A (en) | 2023-07-25 |
| CN116488158B true CN116488158B (en) | 2023-09-15 |
Family
ID=87226398
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310446257.9A Active CN116488158B (en) | 2023-04-24 | 2023-04-24 | Method and device for predicting medium-long-term electric quantity of wind power based on transfer learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116488158B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110533193A (en) * | 2019-08-20 | 2019-12-03 | 武汉理工大学 | Feature and example combine transfer learning method under semi-supervised scene |
| CN113570132A (en) * | 2021-07-23 | 2021-10-29 | 华中科技大学 | Wind power prediction method for space-time meteorological feature extraction and deep learning |
| CN114201924A (en) * | 2022-02-16 | 2022-03-18 | 杭州经纬信息技术股份有限公司 | Solar irradiance prediction method and system based on transfer learning |
| CN114219181A (en) * | 2022-01-20 | 2022-03-22 | 天津大学 | Wind power probability prediction method based on transfer learning |
| CN114493050A (en) * | 2022-04-08 | 2022-05-13 | 南方电网数字电网研究院有限公司 | Multi-dimensional fusion new energy power parallel prediction method and device |
| KR20220107794A (en) * | 2021-01-26 | 2022-08-02 | 한국전력공사 | Power information prediction method and system based on transfer learning |
| CN115511227A (en) * | 2022-11-21 | 2022-12-23 | 国能日新科技股份有限公司 | Wind power generation power prediction method and device based on stable learning |
-
2023
- 2023-04-24 CN CN202310446257.9A patent/CN116488158B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110533193A (en) * | 2019-08-20 | 2019-12-03 | 武汉理工大学 | Feature and example combine transfer learning method under semi-supervised scene |
| KR20220107794A (en) * | 2021-01-26 | 2022-08-02 | 한국전력공사 | Power information prediction method and system based on transfer learning |
| CN113570132A (en) * | 2021-07-23 | 2021-10-29 | 华中科技大学 | Wind power prediction method for space-time meteorological feature extraction and deep learning |
| CN114219181A (en) * | 2022-01-20 | 2022-03-22 | 天津大学 | Wind power probability prediction method based on transfer learning |
| CN114201924A (en) * | 2022-02-16 | 2022-03-18 | 杭州经纬信息技术股份有限公司 | Solar irradiance prediction method and system based on transfer learning |
| CN114493050A (en) * | 2022-04-08 | 2022-05-13 | 南方电网数字电网研究院有限公司 | Multi-dimensional fusion new energy power parallel prediction method and device |
| CN115511227A (en) * | 2022-11-21 | 2022-12-23 | 国能日新科技股份有限公司 | Wind power generation power prediction method and device based on stable learning |
Non-Patent Citations (2)
| Title |
|---|
| 基于迁移学习的无监督跨域人脸表情识别;廖东辉;硕士学位论文-信息科技辑;I138-808 * |
| 基于近似熵的风电功率可预测性研究;杨茂;董骏城;齐;;太阳能学报(第10期);2710-2718 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116488158A (en) | 2023-07-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109902863B (en) | Wind speed prediction method and device based on multi-factor time-space correlation | |
| Ghaderi et al. | Deep forecast: Deep learning-based spatio-temporal forecasting | |
| CN107204615B (en) | Method and system for realizing power prediction | |
| CN113937763B (en) | Wind power prediction method, device, equipment and storage medium | |
| CN105389634A (en) | Combined short-term wind power prediction system and method | |
| CN112149879A (en) | New energy medium-and-long-term electric quantity prediction method considering macroscopic volatility classification | |
| CN113657662B (en) | Downscaling wind power prediction method based on data fusion | |
| CN115345076B (en) | Wind speed correction processing method and device | |
| CN109978242A (en) | The photovoltaic power generation cluster power forecasting method and device of scale are risen based on statistics | |
| CN114462718A (en) | CNN-GRU wind power prediction method based on time sliding window | |
| CN105447572A (en) | Wind power prediction system and method based on neural network optimized by genetic algorithm | |
| CN114792156A (en) | Photovoltaic output power prediction method and system based on curve characteristic index clustering | |
| CN116014722A (en) | Sub-solar photovoltaic power generation prediction method and system based on seasonal decomposition and convolution network | |
| CN115481918A (en) | Active sensing and predictive analysis system for unit state based on source network load storage | |
| CN116316615B (en) | Data enhancement-based distributed light Fu Qun short-term power prediction method and system | |
| CN116488158B (en) | Method and device for predicting medium-long-term electric quantity of wind power based on transfer learning | |
| CN112836876A (en) | Power distribution network line load prediction method based on deep learning | |
| CN115936238B (en) | Method, system, equipment and medium for predicting medium-long-term output of global wind power | |
| CN108694475B (en) | Short-time-scale photovoltaic cell power generation capacity prediction method based on hybrid model | |
| CN116611702A (en) | Integrated learning photovoltaic power generation prediction method for building integrated energy management | |
| CN116454872A (en) | Wind power prediction method and device, electronic equipment and storage medium | |
| CN115329899A (en) | Clustering equivalent model construction method, system, equipment and storage medium | |
| CN114971007A (en) | Photovoltaic power generation prediction method and system based on multi-scale graph convolutional neural network | |
| CN109685242B (en) | Photovoltaic ultra-short term combined prediction method based on Adaboost algorithm | |
| CN113887839A (en) | Clustering-based wind power plant ultra-short-term wind power prediction method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |