CN117725822A - LSTM-based short-term wind power prediction method and system - Google Patents
LSTM-based short-term wind power prediction method and system Download PDFInfo
- Publication number
- CN117725822A CN117725822A CN202311684398.0A CN202311684398A CN117725822A CN 117725822 A CN117725822 A CN 117725822A CN 202311684398 A CN202311684398 A CN 202311684398A CN 117725822 A CN117725822 A CN 117725822A
- Authority
- CN
- China
- Prior art keywords
- prediction
- lstm
- data set
- wind power
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000012549 training Methods 0.000 claims abstract description 56
- 238000013528 artificial neural network Methods 0.000 claims abstract description 12
- 230000007787 long-term memory Effects 0.000 claims abstract description 9
- 230000006403 short-term memory Effects 0.000 claims abstract description 7
- 230000006870 function Effects 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 4
- 238000004422 calculation algorithm Methods 0.000 claims description 3
- 238000011478 gradient descent method Methods 0.000 claims description 3
- 230000007774 longterm Effects 0.000 claims description 3
- 230000015654 memory Effects 0.000 claims description 3
- 210000002569 neuron Anatomy 0.000 claims description 3
- 238000010248 power generation Methods 0.000 abstract description 15
- 238000011161 development Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000007619 statistical method Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007786 learning performance Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012731 temporal analysis Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000000700 time series analysis Methods 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- 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
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a short-term wind power prediction method and a system based on LSTM, comprising the steps of obtaining a historical wind power overall data set of a predicted region; determining a prediction time scale according to the prediction demand, and extracting data of the corresponding time scale from the whole data sets to obtain a prediction data set, wherein the prediction data set comprises a first data set and a second data set; constructing a plurality of data subsets with the same size by utilizing a first data set, constructing a plurality of LSTM training models of the long-term and short-term memory artificial neural network, and training the LSTM training models by utilizing the data subsets to obtain a plurality of different LSTM prediction models; and respectively inputting the second data sets into the LSTM prediction model, and integrating the prediction results of different time scales by adopting a weighted assignment method to obtain a short-term wind power prediction result. The technical scheme of the invention breaks through the short-term wind power generation prediction bottleneck, has high prediction precision and good economy, and effectively supports the reliable grid connection of wind power generation.
Description
Technical Field
The invention relates to the technical field of wind power generation, in particular to a short-term wind power prediction method and system based on LSTM.
Background
The development of new energy has important significance for guaranteeing energy safety, slowing down climate change, promoting economic growth and realizing sustainable development. As one of clean energy sources, wind power can be mutually supplemented with other renewable energy sources (such as solar energy, water energy and the like), so that a more flexible and sustainable energy supply system is formed, dependence on traditional energy sources is reduced, and the safety and the sustainability of the energy sources are improved.
Wind power prediction is particularly important because wind power generation output has volatility. The wind power output prediction method based on the weather data and the historical wind power generation data is a technology for predicting wind power output in a period of time in the future, has important application value, and can help power grid operators, wind power plant main and wind power equipment manufacturers and the like to conduct planning, scheduling, operation optimization, wind power equipment design and other works.
The common wind power prediction method mainly comprises the following steps:
one is prediction based on statistical methods. The method is based on historical data and a statistical model, and uses a statistical method for analysis and prediction. For example, historical wind speed data is used for time series analysis, autoregressive moving average (ARMA), and the like. The method has the advantages of simple calculation and low cost, but the prediction accuracy can be affected under the condition that the wind speed is changed greatly or the nonlinear relation exists.
And secondly, a numerical weather forecast model. And wind power prediction is performed by using a numerical mode and an atmospheric physical equation. The method needs to input various parameters such as atmosphere, geography, topography and the like, and uses a numerical model to carry out numerical simulation and prediction. The method has the advantages of being capable of more accurately simulating the aerodynamic process and having good adaptability to complex terrain and meteorological environments. The disadvantage is that the computation is complex, relatively large amounts of computational resources and data input are required, and for the case of shorter prediction time periods, the prediction accuracy may be lower.
In addition, the real-time monitoring prediction based on sensor data is performed through sensors connected through monitoring equipment, such as an ultrasonic anemometer, a anemoscope and the like, but the cost is high, the accuracy and the reliability of the monitoring equipment are limited, the overall economy is low, and the popularization and application prospect is limited.
In summary, the current prior art has a certain bottleneck in the aspect of short-term wind power generation prediction, mainly shows that the prediction accuracy is low, the economical efficiency is poor, and the like, and is difficult to provide more powerful support for the development of wind power generation.
Disclosure of Invention
In view of the above, the embodiment of the invention provides the LSTM-based short-term wind power prediction method and system, which break through the short-term wind power generation prediction bottleneck, have high prediction precision and good economy, and effectively support the reliable grid connection of wind power generation.
The embodiment of the invention provides a short-term wind power prediction method based on LSTM, which comprises the following steps:
acquiring a historical wind power overall data set of a predicted region;
determining a prediction time scale according to short-term wind power prediction requirements, and extracting data of corresponding time scales from all data sets to obtain a prediction data set, wherein the prediction data set comprises a first data set and a second data set;
constructing a plurality of data subsets with the same size by utilizing the first data set, constructing a plurality of LSTM training models of the long-short-term memory artificial neural network, and training the LSTM training models by utilizing the data subsets to obtain a plurality of different LSTM prediction models;
and respectively inputting the second data set into the plurality of different LSTM prediction models to respectively obtain a plurality of prediction results of different time scales, and integrating the prediction results of different time scales by adopting a weighted assignment method to obtain a short-term wind power prediction result.
Illustratively, the historical wind power ensemble data set includes generated power, wind speed, wind direction, and temperature information, with a sampling frequency of 1h for 24 data points throughout the day.
Illustratively, the first data set is used for model training, the second data set is used for power prediction, and the sample composition time scales of the plurality of data subsets are different.
Illustratively, constructing a plurality of data subsets of the same size using the first data set, and constructing a plurality of long-term memory artificial neural network LSTM training models includes:
setting the length of a data set input sequence and a time step length, wherein the length of the data set input sequence is the size of a time window, and the time step length is the time interval between samples;
the LSTM training model is constructed, the LSTM training model comprises an input layer, an LSTM layer and an output layer, the input layer is used for receiving the first data set, the LSTM layer is used for solving long-term dependence in the data set, and the output layer is used for predicting short-term wind power;
model parameters are set, and the model parameters comprise the quantity of neurons of the LSTM layer and an optimizer.
Illustratively, the training the LSTM training model with the subset of data to obtain a plurality of different LSTM prediction models includes:
inputting the second dataset into the LSTM training model;
the model parameters are optimized using a back-propagation algorithm and a gradient descent method to minimize the loss function between the predicted and actual values.
The step of respectively inputting the second data set to the plurality of different LSTM prediction models to respectively obtain a plurality of prediction results of different time scales, and the step of integrating the prediction results of different time scales by using a weighted assignment method to obtain a short-term wind power prediction result includes:
respectively inputting the second data set into the plurality of different LSTM prediction models, wherein the prediction time scales of the LSTM prediction models are respectively 1h, 2h, 5h and 10h, and obtaining prediction results based on different time scales;
and integrating the prediction results of different time scales by adopting a weighted average method to obtain a final prediction result.
Illustratively, the method further comprises:
the prediction performance is evaluated using the root mean square error and the average absolute error.
Another embodiment of the present invention proposes an LSTM based short-term wind power prediction system comprising:
the data acquisition unit is used for acquiring a historical wind power overall data set of the predicted region;
the data set construction unit is used for determining a prediction time scale according to short-term wind power prediction requirements, and extracting data of the corresponding time scale from the whole data sets to obtain a prediction data set, wherein the prediction data set comprises a first data set and a second data set;
the model training unit is used for constructing a plurality of data subsets with the same size by utilizing the first data set, constructing a plurality of LSTM training models of the long-term and short-term memory artificial neural network, and training the LSTM training models by utilizing the data subsets to obtain a plurality of different LSTM prediction models;
and the prediction unit is used for respectively inputting the second data set into the plurality of different LSTM prediction models to respectively obtain a plurality of prediction results of different time scales, and integrating the prediction results of different time scales by adopting a weighted assignment method to obtain a short-term wind power prediction result.
The invention provides a short-term wind power prediction method and a system based on LSTM, comprising the steps of obtaining a historical wind power overall data set of a predicted region; determining a prediction time scale according to the prediction demand, and extracting data of the corresponding time scale from the whole data sets to obtain a prediction data set, wherein the prediction data set comprises a first data set and a second data set; constructing a plurality of data subsets with the same size by utilizing a first data set, constructing a plurality of LSTM training models of the long-term and short-term memory artificial neural network, and training the LSTM training models by utilizing the data subsets to obtain a plurality of different LSTM prediction models; and respectively inputting the second data sets into the LSTM prediction model, and integrating the prediction results of different time scales by adopting a weighted assignment method to obtain a short-term wind power prediction result. The technical scheme of the invention breaks through the short-term wind power generation prediction bottleneck, has high prediction precision and good economy, and effectively supports the reliable grid connection of wind power generation.
Drawings
In order to more clearly illustrate the technical solutions of the present invention, the drawings that are required for the embodiments will be briefly described, it being understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope of the present invention. Like elements are numbered alike in the various figures.
FIG. 1 is a schematic flow chart of a short-term wind power prediction method based on LSTM provided by the invention;
FIG. 2 is a flow chart of the method of step S103 provided by the present invention;
FIG. 3 is a schematic flow chart of another step S103 method provided by the present invention;
FIG. 4 is a flow chart of the step S104 method provided by the present invention;
FIG. 5 is a schematic diagram of a short-term wind power prediction system based on LSTM.
Description of main reference numerals:
10-a data acquisition unit; 20-a dataset construction unit; 30-a model training unit; 40-prediction unit.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments.
Example 1
Wind energy is a clean energy source, the installed capacity of wind power is gradually increased every year at present, however, due to the fact that wind power generation is affected by randomness of weather environment, certain output fluctuation exists, the wind abandoning rate exists objectively, and the wind abandoning rate is also an influence factor of electric power instability of a power grid. In this regard, the embodiment of the invention provides the short-term wind power prediction method based on LSTM, thereby realizing accurate prediction of short-term wind power, having good prediction precision and the prediction accuracy rate reaching 95.8%.
Referring to fig. 1, the short-term wind power prediction method based on LSTM includes:
step S101, acquiring a historical wind power overall data set of a predicted region;
specifically, the historical wind power total data set comprises power generation power, wind speed, wind direction and temperature information, the sampling frequency is 1h, and 24 data points are all obtained in the whole day. In particular, because the method is limited by an on-site acquisition environment, only small sample data can be obtained generally, and the information is limited. The method is realized by mainly sampling a plurality of sub-data sets from a training data set by self-help sampling with a place-back way, then training a base learner for each sub-data set, and finally determining a final prediction result by voting or averaging and the like. The prediction results are combined to improve the overall learning performance by constructing a plurality of mutually independent base learners.
Step S102, determining a prediction time scale according to short-term wind power prediction requirements, and extracting data of the corresponding time scale from all data sets to obtain a prediction data set, wherein the prediction data set comprises a first data set and a second data set;
in particular, the first data set is used for model training, the second data set is used for power prediction, and the sample composition time scales of the plurality of data subsets are different. Here, the first data set is typically a training set and the second data set is a test set.
Step S103, constructing a plurality of data subsets with the same size by using the first data set, constructing a plurality of LSTM training models of the long-short-term memory artificial neural network, and training the LSTM training models by using the data subsets to obtain a plurality of different LSTM prediction models;
step S104, the second data set is respectively input into a plurality of different LSTM prediction models to respectively obtain a plurality of prediction results of different time scales, and the prediction results of different time scales are integrated by adopting a weighted assignment method to obtain a short-term wind power prediction result.
Referring to fig. 2, constructing a plurality of data subsets with the same size using the first data set and constructing a plurality of long-term memory artificial neural network LSTM training models in step S103 includes:
step S201, setting the length of a data set input sequence and a time step length, wherein the length of the data set input sequence is the size of a time window, and the time step length is the time interval between samples;
step S202, constructing an LSTM training model, wherein the LSTM training model comprises an input layer, an LSTM layer and an output layer, the input layer is used for receiving a first data set, the LSTM layer is used for solving long-term dependency relationship in the data set, and the output layer is used for predicting short-term wind power;
in step S203, model parameters are set, where the model parameters include the number of neurons of the LSTM layer and the optimizer.
Specifically, the LSTM model provided by the embodiment of the invention can automatically judge which information needs to be forgotten from the data, realize the forgetting of the related information through a forgetting gate, and judge the ratio of the information forgotten at the last moment by utilizing a sigmoid function; then, selectively storing the new input information obtained through the input gate, and determining new information to be recorded by utilizing a sigmoid function; and finally, realizing information output through an output gate, calculating the output information proportion by using a sigmoid function, and carrying out series operation to obtain a final output result.
Referring to fig. 3, training the LSTM training model using the subset of data to obtain a plurality of different LSTM prediction models in step S103 includes:
step S301, inputting a second data set into an LSTM training model;
in step S302, model parameters are optimized using a back-propagation algorithm and a gradient descent method to minimize a loss function between the predicted value and the actual value.
Specifically, the embodiment of the invention adopts a strategy of training a plurality of models, a plurality of scale time series data are selected to form a diversified data set, at least 2 LSTM training models are trained and predicted, and as the influence of the data with different scales on the prediction time points is different, the embodiment of the invention can give different weights, so that the combination of model prediction results is completed, and a final prediction result is formed.
Referring to fig. 4, step S104 includes:
step S401, respectively inputting a second data set into a plurality of different LSTM prediction models, wherein the prediction time scales of the LSTM prediction models are respectively 1h, 2h, 5h and 10h, and obtaining prediction results based on different time scales;
step S402, integrating the prediction results of different time scales by adopting a weighted average method to obtain a final prediction result;
step S403, the prediction performance is evaluated using the root mean square error and the average absolute error.
Specifically, the embodiment of the invention can apply the weighted voting rule to integrate the model prediction results.
The embodiment of the invention discloses a short-term wind power prediction method and a short-term wind power prediction system based on LSTM, which comprise the steps of acquiring a historical wind power overall data set of a predicted region; determining a prediction time scale according to the prediction demand, and extracting data of the corresponding time scale from the whole data sets to obtain a prediction data set, wherein the prediction data set comprises a first data set and a second data set; constructing a plurality of data subsets with the same size by utilizing a first data set, constructing a plurality of LSTM training models of the long-term and short-term memory artificial neural network, and training the LSTM training models by utilizing the data subsets to obtain a plurality of different LSTM prediction models; and respectively inputting the second data sets into the LSTM prediction model, and integrating the prediction results of different time scales by adopting a weighted assignment method to obtain a short-term wind power prediction result. The technical scheme of the invention breaks through the short-term wind power generation prediction bottleneck, has high prediction precision and good economy, and effectively supports the reliable grid connection of wind power generation.
Example 2
Referring to FIG. 5, an embodiment of the present invention further provides an LSTM-based short-term wind power prediction system, comprising:
a data acquisition unit 10 for acquiring a historical wind power total data set of a predicted region;
a data set construction unit 20, configured to determine a prediction time scale according to a short-term wind power prediction requirement, and extract data of a corresponding time scale from the whole data set to obtain a prediction data set, where the prediction data set includes a first data set and a second data set;
the model training unit 30 is configured to construct a plurality of data subsets with the same size by using the first data set, construct a plurality of long-term and short-term memory artificial neural network LSTM training models, and train the LSTM training models by using the data subsets to obtain a plurality of different LSTM prediction models;
the prediction unit 40 is configured to input the second data set to a plurality of different LSTM prediction models, respectively, obtain a plurality of prediction results of different time scales, and integrate the prediction results of different time scales by using a weighted assignment method to obtain a short-term wind power prediction result.
It will be appreciated that the LSTM-based short-term wind power prediction system of the above embodiment 2 corresponds to the LSTM-based short-term wind power prediction method of embodiment 1. Any of the alternatives in embodiment 1 are also applicable to this embodiment and will not be described in detail here.
The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present invention.
Claims (8)
1. An LSTM-based short-term wind power prediction method, comprising:
acquiring a historical wind power overall data set of a predicted region;
determining a prediction time scale according to short-term wind power prediction requirements, and extracting data of corresponding time scales from all data sets to obtain a prediction data set, wherein the prediction data set comprises a first data set and a second data set;
constructing a plurality of data subsets with the same size by utilizing the first data set, constructing a plurality of LSTM training models of the long-short-term memory artificial neural network, and training the LSTM training models by utilizing the data subsets to obtain a plurality of different LSTM prediction models;
and respectively inputting the second data set into the plurality of different LSTM prediction models to respectively obtain a plurality of prediction results of different time scales, and integrating the prediction results of different time scales by adopting a weighted assignment method to obtain a short-term wind power prediction result.
2. The LSTM based short-term wind power prediction method of claim 1, wherein the historical wind power ensemble data set includes generated power, wind speed, wind direction, and temperature information, with a sampling frequency of 1h for 24 data points throughout the day.
3. The LSTM based short-term wind power prediction method of claim 2 wherein the first data set is used for model training and the second data set is used for power prediction, the sample composition time scales of the plurality of data subsets being different.
4. The LSTM based short-term wind power prediction method of claim 3, wherein constructing a plurality of data subsets of the same size using the first data set, and constructing a plurality of long-term and short-term memory artificial neural network LSTM training models includes:
setting the length of a data set input sequence and a time step length, wherein the length of the data set input sequence is the size of a time window, and the time step length is the time interval between samples;
the LSTM training model is constructed, the LSTM training model comprises an input layer, an LSTM layer and an output layer, the input layer is used for receiving the first data set, the LSTM layer is used for solving long-term dependence in the data set, and the output layer is used for predicting short-term wind power;
model parameters are set, and the model parameters comprise the quantity of neurons of the LSTM layer and an optimizer.
5. The LSTM based short-term wind power prediction method of claim 4, wherein training the LSTM training model using the subset of data to obtain a plurality of different LSTM prediction models comprises:
inputting the second dataset into the LSTM training model;
the model parameters are optimized using a back-propagation algorithm and a gradient descent method to minimize the loss function between the predicted and actual values.
6. The LSTM-based short-term wind power prediction method of claim 5, wherein the inputting the second data set into the plurality of different LSTM prediction models to obtain a plurality of prediction results of different time scales, and integrating the prediction results of different time scales by using a weighted assignment method to obtain the short-term wind power prediction result includes:
respectively inputting the second data set into the plurality of different LSTM prediction models, wherein the prediction time scales of the LSTM prediction models are respectively 1h, 2h, 5h and 10h, and obtaining prediction results based on different time scales;
and integrating the prediction results of different time scales by adopting a weighted average method to obtain a final prediction result.
7. The LSTM based short-term wind power prediction method of claim 6, further comprising:
the prediction performance is evaluated using the root mean square error and the average absolute error.
8. An LSTM based short-term wind power prediction system comprising:
the data acquisition unit is used for acquiring a historical wind power overall data set of the predicted region;
the data set construction unit is used for determining a prediction time scale according to short-term wind power prediction requirements, and extracting data of the corresponding time scale from the whole data sets to obtain a prediction data set, wherein the prediction data set comprises a first data set and a second data set;
the model training unit is used for constructing a plurality of data subsets with the same size by utilizing the first data set, constructing a plurality of LSTM training models of the long-term and short-term memory artificial neural network, and training the LSTM training models by utilizing the data subsets to obtain a plurality of different LSTM prediction models;
and the prediction unit is used for respectively inputting the second data set into the plurality of different LSTM prediction models to respectively obtain a plurality of prediction results of different time scales, and integrating the prediction results of different time scales by adopting a weighted assignment method to obtain a short-term wind power prediction result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311684398.0A CN117725822A (en) | 2023-12-08 | 2023-12-08 | LSTM-based short-term wind power prediction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311684398.0A CN117725822A (en) | 2023-12-08 | 2023-12-08 | LSTM-based short-term wind power prediction method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117725822A true CN117725822A (en) | 2024-03-19 |
Family
ID=90200869
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311684398.0A Pending CN117725822A (en) | 2023-12-08 | 2023-12-08 | LSTM-based short-term wind power prediction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117725822A (en) |
-
2023
- 2023-12-08 CN CN202311684398.0A patent/CN117725822A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103390116B (en) | Use the photovoltaic power station power generation power forecasting method of stepping mode | |
| CN110909919A (en) | Photovoltaic power prediction method of depth neural network model with attention mechanism fused | |
| CN102411729B (en) | Wind power prediction method based on adaptive linear logic network | |
| CN110110912B (en) | Photovoltaic power multi-model interval prediction method | |
| CN106228278A (en) | Photovoltaic power prognoses system | |
| CN103268366A (en) | Combined wind power prediction method suitable for distributed wind power plant | |
| CN108388962A (en) | A kind of wind power forecasting system and method | |
| CN113496311A (en) | Photovoltaic power station generated power prediction method and system | |
| CN111695736B (en) | Photovoltaic power generation short-term power prediction method based on multi-model fusion | |
| CN106875033A (en) | A kind of wind-powered electricity generation cluster power forecasting method based on dynamic self-adapting | |
| CN108196317B (en) | Meteorological prediction method for micro-grid system | |
| CN115481918A (en) | Active sensing and predictive analysis system for unit state based on source network load storage | |
| CN114021848A (en) | Generating capacity demand prediction method based on LSTM deep learning | |
| CN114021830A (en) | Multi-time-range wind speed prediction method based on CNN-LSTM | |
| CN107358059A (en) | Short-term photovoltaic energy Forecasting Methodology and device | |
| CN114895380A (en) | Solar radiation prediction method, device, equipment and medium | |
| CN108710966B (en) | Photovoltaic power generation power prediction method based on multi-cluster ESN neural network | |
| CN104008284A (en) | Correcting method for anemometer tower in numerical weather prediction | |
| CN105958474B (en) | Dynamic capacity increasing method and system for power transmission line for power grid regulation and control system | |
| CN116611702A (en) | Integrated learning photovoltaic power generation prediction method for building integrated energy management | |
| Potter et al. | Wind power data for grid integration studies | |
| CN111177278A (en) | Grid user short-term load prediction real-time processing tool | |
| CN117725822A (en) | LSTM-based short-term wind power prediction method and system | |
| CN115912334A (en) | Method for establishing prediction model of output guarantee rate of wind power plant and prediction method | |
| CN110705769B (en) | New energy power generation power prediction optimization method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication |