CN115511227B - Wind power generation power prediction method and device based on stable learning - Google Patents
Wind power generation power prediction method and device based on stable learning Download PDFInfo
- Publication number
- CN115511227B CN115511227B CN202211452304.2A CN202211452304A CN115511227B CN 115511227 B CN115511227 B CN 115511227B CN 202211452304 A CN202211452304 A CN 202211452304A CN 115511227 B CN115511227 B CN 115511227B
- Authority
- CN
- China
- Prior art keywords
- data
- sample
- training set
- model
- wind power
- 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
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000010248 power generation Methods 0.000 title claims abstract description 20
- 238000012549 training Methods 0.000 claims abstract description 59
- 239000013598 vector Substances 0.000 claims abstract description 37
- 238000012360 testing method Methods 0.000 claims abstract description 8
- 238000012795 verification Methods 0.000 claims abstract description 6
- 230000004927 fusion Effects 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000013145 classification model Methods 0.000 claims description 7
- 238000012805 post-processing Methods 0.000 claims description 7
- 238000013461 design Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000012417 linear regression Methods 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Supply And Distribution Of Alternating Current (AREA)
- Wind Motors (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a wind power generation power prediction method and device based on stable learning, wherein a data set for constructing a wind power plant power prediction model is divided into a training set, a verification set and a test set according to a time stamp; constructing a characteristic vector from historical wind speed data in training set data; calculating distance variables between the feature vectors and calculating a distance between the feature vectors based on the distance variables s i,j Carrying out feature clustering; learning the training set data samples to sample weights through a stable learning algorithm according to the feature clustering result; and constructing a wind power plant power prediction model by using the weighted samples. The method and the device for predicting the wind power generation power based on the stable learning provide a way for solving the OOD problem and ensure the stability of the prediction of the wind power generation power.
Description
Technical Field
The invention belongs to the technical field of new energy electric power, and particularly relates to a wind power generation power prediction method and device based on stable learning.
Background
The wind power plant power prediction is generally that a prediction model of wind power output power is established according to historical power, historical wind speed, landform and numerical weather forecast operation states of a wind power plant, wind speed, power or numerical weather forecast data are used as model input, modeling is carried out through a machine learning algorithm, and time series prediction is carried out on the power. In order to ensure the effect of the prediction model, the machine learning algorithm requires that training set data and test set data are subjected to independent equal distribution (i.i.d.), but actually, the requirement is very strict, in an actual scene, i.i.d. assumption is difficult to meet, so that the performance of the classical machine learning algorithm under the distribution change is sharply reduced, the problem of out-of-distribution generalization (OOD) is generated, and finally the instability of wind power generation power prediction is caused.
Disclosure of Invention
The invention provides a wind power generation power prediction method and device based on stable learning, thereby providing a way for solving an OOD problem and ensuring the stability of wind power generation power prediction.
In order to achieve the purpose, the technical scheme of the invention is realized as follows:
a wind power generation power prediction method based on stable learning comprises the following steps:
s1, taking historical power, historical wind speed and numerical weather forecast historical data of a wind power plant as a data set for constructing a wind power plant power prediction model, and dividing the data set into a training set, a verification set and a test set according to a timestamp;
s2, constructing a characteristic vector from historical wind speed data in the training set data;
s3, calculating distance variables between feature vectors
(ii) a Wherein->
；x i,j Is a feature vector x i ,x j Is based on the distance characteristic value of (4), is based on the value of (4)>
Is a feature vector x i ,x j The cosine of the distance similarity of (a), device for selecting or keeping>
Is a feature vector x i ,x j The Euclidean distance of (c);
according to the distance variable s between the feature vectors i,j Carrying out feature clustering;
s4, integrating to obtain a clustered training set according to the characteristic clustering result; learning the data samples of the clustered training set to sample weights through a stable learning algorithm, and weighting the data samples; and constructing a wind power plant power prediction model by using the weighted training set data samples.
Further, the specific process of learning the data samples of the clustered training set to the sample weight through the stable learning algorithm in step S4 includes:
s401, generating a data set Y through the clustered training set X: the generation process is to randomly extract the column characteristics of X and recombine to obtain a data set Y;
s402, setting the sample label z of the clustered training set X as 1 and setting the sample label z of the data set Y as 0;
s403, merging the clustered training set X and the clustered data set Y, and then training a two-classification model;
s404, calculating each sample X in the clustered training set X
W (x) is the sample weight of x; />
A probability that the label z for a sample x equals 1, based on the value of>
Probability of label z being sample x being equal to 0;
and S405, weighting the data samples according to the sample weights for modeling.
Further, the wind power plant power prediction model in the step S4 comprises fusion of a linear model and a tree model; the fusion mode comprises bagging and/or boosting, wherein the bagging is to perform weighted combination on the result of the linear model and the result of the tree model, and the boosting is to take the output of the linear model as the input of the tree model.
Furthermore, after a wind power plant power prediction model is built, post-processing is carried out on prediction data according to the power curve parameters of the wind turbine during prediction.
The invention also provides a wind power generation power prediction device based on stable learning, which comprises:
a data set module: taking historical power, historical wind speed and numerical weather forecast historical data of a wind power plant as a data set for constructing a wind power plant power prediction model, and dividing the data set into a training set, a verification set and a test set according to a timestamp;
a feature vector module: constructing a characteristic vector by using historical wind speed data in training set data;
a clustering module: between calculation of feature vectorsVariable of distance
(ii) a Wherein->
；x i,j Is a feature vector x i ,x j In a distance characteristic value of (a), in a manner that>
Is a feature vector x i ,x j Based on the cosine distance similarity, and->
Is a feature vector x i ,x j The Euclidean distance of (c); according to the distance variable s between the feature vectors i,j Carrying out feature clustering;
a weighting module: integrating to obtain a clustered training set according to the characteristic clustering result; learning the data samples of the clustered training set to sample weights through a stable learning algorithm, and weighting the data samples; and constructing a wind power plant power prediction model by using the weighted training set data samples.
Further, the weighting module includes:
a data set unit: generating a data set Y through the clustered training set X: the generation process is to randomly extract the column characteristics of X and recombine to obtain a data set Y;
a sample label unit: the sample label z of the clustered training set X is 1, and the sample label z of the data set Y is 0;
a two-classification model training unit: merging the clustered training set X and the clustered data set Y, and then training a two-classification model;
a weight calculation unit: calculating each sample X in the clustered training set X
W (x) is the sample weight of x; />
A probability that the label z for a sample x equals 1, based on the value of>
A probability that label z for sample x equals 0;
a weighting unit: and weighting the data samples according to the sample weights for modeling.
Further, the wind power plant power prediction model in the weighting module comprises fusion of a linear model and a tree model; the fusion mode comprises bagging and/or boosting, wherein the bagging is to perform weighted combination on the result of the linear model and the result of the tree model, and the boosting is to take the output of the linear model as the input of the tree model.
The system further comprises a post-processing module, wherein the post-processing module is used for performing post-processing on the prediction data according to the power curve parameters of the wind turbine during prediction after a wind power plant power prediction model is built.
Compared with the prior art, the invention has the following beneficial effects:
1. the invention provides a distance variable calculation method, which is used for carrying out feature clustering on a training set, reducing the collinearity of features and solving the problem of out-of-distribution generalization (OOD);
2. according to the method, the samples are weighted through a stable learning algorithm on the basis of carrying out feature clustering on the training set, and the method is used for further solving the problem of out-of-distribution generalization (OOD) and ensuring the stability of wind power generation power prediction.
Drawings
FIG. 1 is a schematic flow diagram of an embodiment of the present invention;
fig. 2 is a schematic flow chart of the stable learning algorithm in the embodiment of the present invention.
Detailed Description
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict.
The design idea of the invention is to solve the OOD problem by sample weighting to reduce the co-linearity of features.
Based on the above design concept, the invention provides a wind power generation power prediction method based on stable learning, as shown in fig. 1, specifically comprising:
s1, taking historical power, historical wind speed and numerical weather forecast historical data of a wind power plant as a data set for constructing a wind power plant power prediction model, and dividing the data set into a training set, a verification set and a test set according to a timestamp; and carrying out abnormal value processing: including null, dead, over-limit, etc
S2, performing feature processing on the historical wind speed data in the training set data through feature engineering to construct feature vectors; the characteristic processing is mainly statistical characteristics and also comprises the step of taking period transformation sin (), namely taking the date;
s3, calculating distance variables between feature vectors
(ii) a Wherein->
；x i,j Is a feature vector x i ,x j Is based on the distance characteristic value of (4), is based on the value of (4)>
Is a feature vector x i ,x j The cosine distance similarity of (a) is, device for combining or screening>
Is a feature vector x i ,x j The Euclidean distance of (c);
feature vector x i ,x j The cosine distance similarity calculation method comprises the following steps:
；
feature vector x i ,x j The calculated orientation of the euclidean distance of (c) is:
(ii) a k is the number of elements of the feature vector;
in the present invention, the distance variable s is i,j The design of (c) is as follows: distance variable s i,j Is to use the distance characteristic value x i,j Obtained by substituting sigmoid function; while the distance eigenvalue x i,j The similarity of distance between the numerator and the cosine in the calculation formula
The cosine distance similarity means an included angle of two eigenvectors, which are different in height in space but have the same included angle and a large difference in weather; in addition at a distance characteristic value x i,j The denominator in the calculation formula applies the Euclidean distance between two eigenvectors, so that the distance between the two eigenvectors with the same included angle but different heights is different, the larger the distance between the two points is, the smaller the collinearity is, and finally, the distance eigenvalue x is obtained i,j Is designed to be->
(ii) a So as to adjust the distance characteristic value x i,j The substitution into the sigmoid function is because the actual distance size has no meaning after the distance is large to some extent.
Calculating a distance variable s between feature vectors i,j Then, according to the distance variable s i,j Carrying out feature clustering; feature clustering may use a K-means clustering method.
S4, integrating the training set according to the characteristic clustering result to obtain a clustered training set, and learning the data sample of the clustered training set to the sample weight through a Stable learning algorithm (Stable learning); the stable learning algorithm starts from collinearity (collinearity), which aggravates the error of parameter estimation and leads to instability of prediction. The sample weight w (x) is learned through a stable learning algorithm, and the weighted sample is used for constructing a wind power plant power prediction model.
The specific process of stable learning is shown in fig. 2, and includes:
s401, setting the clustered training set as a data set X, and generating a data set Y through the data set X: the generation process is to randomly extract the column characteristics of X and recombine to obtain a data set Y;
s402, setting a sample label z of a data set X as 1 and setting a sample label z of a data set Y as 0;
s403, merging the clustered data set X and the clustered data set Y, and then training a two-classification model;
s404, calculating each sample X in the data set X
W (x) is the sample weight of x;
a probability that the label z for a sample x equals 1, based on the value of>
Probability of label z being sample x being equal to 0;
s405, weighting the data samples of the clustered training set according to the sample weights, wherein the weighting is specifically that a characteristic coefficient is calculated by a linear model; and the weighted samples are used for constructing a wind power plant power prediction model.
The weighted samples are used for constructing the wind power plant power prediction model, the wind power plant power prediction model in the embodiment of the invention comprises fusion of a linear regression model and a gbm tree model, the fusion mode comprises a bagging method and/or a boosting method, the bagging method is that a result of the linear regression model and a result of the tree model are combined in a weighted mode, and the boosting method is that the output of the linear regression model is used as the input of the tree model.
After the wind power plant power prediction model is built, the prediction data is subjected to post-processing according to the power curve parameters of the wind turbine. The specific post-treatment process can be as follows: the curve is divided into 3 sections according to the inflection point of the power curve of the fan, and each section is optimized according to a loss function.
Aiming at the problem of high collinearity of data features in a data set, the invention provides a distance variable calculation method which is used for carrying out feature clustering on a training set and reducing the collinearity among features, and also carrying out weighting on samples through a stable learning algorithm, thereby solving the problem of out-of-distribution (OOD), reducing the data difference of the training set and a test set and obeying independent same distribution; and the stability of the wind power generation power prediction is ensured.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (6)
1. A wind power generation power prediction method based on stable learning is characterized by comprising the following steps:
s1, taking historical power, historical wind speed and numerical weather forecast historical data of a wind power plant as a data set for constructing a wind power plant power prediction model, and dividing the data set into a training set, a verification set and a test set according to a timestamp;
s2, constructing a characteristic vector from historical wind speed data in the training set data;
s3, calculating distance variables between feature vectors
(ii) a Wherein->
；x i,j Is a feature vector x i ,x j Is based on the distance characteristic value of (4), is based on the value of (4)>
Is a feature vector x i ,x j Based on the cosine distance similarity, and->
Is a feature vector x i ,x j The Euclidean distance of (c);
according to the distance variable s between the feature vectors i,j Carrying out feature clustering;
s4, integrating to obtain a clustered training set according to the characteristic clustering result; learning the data samples of the clustered training set to sample weights through a stable learning algorithm, and weighting the data samples; constructing a wind power plant power prediction model by using the weighted training set data samples;
the specific process of learning the data samples of the clustered training set to the sample weight through the stable learning algorithm in the step S4 includes:
s401, setting the clustered training set as a data set X, and generating a data set Y through the data set X: the generation process is to randomly extract the column characteristics of X and recombine to obtain a data set Y;
s402, setting a sample label z of a data set X as 1 and setting a sample label z of a data set Y as 0;
s403, merging the clustered data set X and the clustered data set Y, and then training a two-classification model;
s404, calculating each sample X in the data set X
W (x) is the sample weight of x; />
A probability that the label z for a sample x equals 1, based on the value of>
Probability of label z being sample x being equal to 0;
s405, weighting the data samples of the clustered training set according to the sample weights, wherein the weighting is specifically that a characteristic coefficient is calculated by a linear model; and the weighted samples are used for constructing a wind power plant power prediction model.
2. The stable learning based wind power generation power prediction method according to claim 1, characterized in that the wind farm power prediction model in step S4 comprises a fusion of a linear model and a tree model; the fusion mode comprises bagging and/or boosting, wherein the bagging is to perform weighted combination on the result of the linear model and the result of the tree model, and the boosting is to take the output of the linear model as the input of the tree model.
3. The wind power generation power prediction method based on stable learning of claim 1, wherein after the wind power plant power prediction model is constructed, prediction data is post-processed according to the power curve parameters of the wind turbine during prediction.
4. A wind power generation power prediction device based on stable learning is characterized by comprising:
a data set module: taking historical power, historical wind speed and numerical weather forecast historical data of a wind power plant as a data set for constructing a power prediction model of the wind power plant, and dividing the data set into a training set, a verification set and a test set according to a time stamp;
a feature vector module: constructing a characteristic vector from historical wind speed data in training set data;
a clustering module: computing distance variables between feature vectors
(ii) a Wherein->
；x i,j Is a feature vector x i ,x j Is based on the distance characteristic value of (4), is based on the value of (4)>
Is a feature vector x i ,x j Based on the cosine distance similarity, and->
Is a feature vector x i ,x j The Euclidean distance of (c); according to the distance variable s between the feature vectors i,j Carrying out feature clustering;
a weighting module: integrating to obtain a clustered training set according to the characteristic clustering result; learning the data samples of the clustered training set to sample weights through a stable learning algorithm, and weighting the data samples; constructing a wind power plant power prediction model by using the weighted training set data samples;
the weighting module includes:
a data set unit: and (3) setting the clustered training set as a data set X, and generating a data set Y through the data set X: the generation process is to randomly extract the column characteristics of X and recombine to obtain a data set Y;
sample label unit: the sample label z of the clustered training set X is 1, and the sample label z of the data set Y is 0;
a two-classification model training unit: merging the clustered training set X and the clustered data set Y, and then training a two-classification model;
a weight calculation unit: calculating each sample X in the clustered training set X
W (x) is the sample weight of x; />
A probability that the label z for a sample x equals 1, based on the value of>
A probability that label z for sample x equals 0;
a weighting unit: weighting the data samples of the clustered training set according to the sample weights, wherein the weighting specifically comprises calculating characteristic coefficients by a linear model; and the weighted samples are used for constructing a wind power plant power prediction model.
5. The stable learning based wind power generation power prediction device according to claim 4, wherein the wind farm power prediction model in the weighting module comprises a fusion of a linear model and a tree model; the fusion mode comprises bagging and/or boosting, wherein the bagging is to perform weighted combination on the result of the linear model and the result of the tree model, and the boosting is to take the output of the linear model as the input of the tree model.
6. The wind power generation power prediction device based on stable learning of claim 4 further comprising a post-processing module for post-processing the prediction data according to the power curve parameters of the wind turbine during prediction after a wind farm power prediction model is constructed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211452304.2A CN115511227B (en) | 2022-11-21 | 2022-11-21 | Wind power generation power prediction method and device based on stable learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211452304.2A CN115511227B (en) | 2022-11-21 | 2022-11-21 | Wind power generation power prediction method and device based on stable learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115511227A CN115511227A (en) | 2022-12-23 |
| CN115511227B true CN115511227B (en) | 2023-04-07 |
Family
ID=84513705
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211452304.2A Active CN115511227B (en) | 2022-11-21 | 2022-11-21 | Wind power generation power prediction method and device based on stable learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115511227B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116488158B (en) * | 2023-04-24 | 2023-09-15 | 国能日新科技股份有限公司 | Method and device for predicting medium-long-term electric quantity of wind power based on transfer learning |
| CN116826731B (en) * | 2023-06-30 | 2024-08-20 | 昆明理工大学 | Self-adaptive wind power prediction method and system based on continuous learning |
| CN117195093B (en) * | 2023-09-20 | 2024-07-30 | 长江水利委员会长江科学院 | Hydropower station water consumption rate calculation method based on KNN machine learning algorithm |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110363354B (en) * | 2019-07-16 | 2023-05-02 | 上海交通大学 | Wind power prediction method for wind farm, electronic device and storage medium |
| CN112288157A (en) * | 2020-10-27 | 2021-01-29 | 华能酒泉风电有限责任公司 | Wind power plant power prediction method based on fuzzy clustering and deep reinforcement learning |
| CN114970698B (en) * | 2022-05-17 | 2024-06-18 | 河海大学 | Metering equipment operation performance prediction method based on improved LWPS |
-
2022
- 2022-11-21 CN CN202211452304.2A patent/CN115511227B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN115511227A (en) | 2022-12-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115511227B (en) | Wind power generation power prediction method and device based on stable learning | |
| Zhang et al. | Local parameter optimization of LSSVM for industrial soft sensing with big data and cloud implementation | |
| CN111091236B (en) | Multi-classification deep learning short-term wind power prediction method classified according to pitch angles | |
| CN109145948A (en) | A kind of injection molding machine putty method for detecting abnormality based on integrated study | |
| CN113822418A (en) | Wind power plant power prediction method, system, device and storage medium | |
| CN110991737A (en) | Ultra-short-term wind power prediction method based on deep belief network | |
| CN111274532B (en) | Short-term wind power prediction method and device based on CEEMD-LZC and manifold learning | |
| CN117874639B (en) | Mechanical equipment service life prediction method and device based on artificial intelligence | |
| CN115689008A (en) | CNN-BilSTM short-term photovoltaic power prediction method and system based on ensemble empirical mode decomposition | |
| Han et al. | Multi-step short-term wind power prediction based on three-level decomposition and improved grey wolf optimization | |
| CN113807568B (en) | Power load prediction method and device and terminal equipment | |
| CN114021483A (en) | Ultra-short-term wind power prediction method based on time domain characteristics and XGboost | |
| Anh et al. | A benchmark of deep learning models for multi-leaf diseases for edge devices | |
| CN114418180A (en) | Ultra-short-term prediction method and device for wind power and storage medium | |
| CN115392595A (en) | Time-space short-term wind speed prediction method and system based on graph convolution neural network and Transformer | |
| CN118114955A (en) | Power scheduling method of virtual power plant and related equipment | |
| CN114021818A (en) | Wind power multistep prediction method considering space-time distribution characteristics | |
| CN114118401A (en) | Neural network-based power distribution network flow prediction method, system, device and storage medium | |
| CN114298132A (en) | Wind power prediction method and device and electronic equipment | |
| CN116245259B (en) | Photovoltaic power generation prediction method and device based on depth feature selection and electronic equipment | |
| CN108038518A (en) | A kind of photovoltaic generation power based on meteorological data determines method and system | |
| CN112115984A (en) | Tea garden abnormal data correction method and system based on deep learning and storage medium | |
| Lyu et al. | A data-driven solar irradiance forecasting model with minimum data | |
| CN115130743A (en) | Wind turbine generator regulation rate prediction method and system based on variation inference | |
| CN114168822A (en) | Method for establishing time series data clustering model and time series data clustering |
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 |