CN114483485B - Method for improving wind speed prediction of Nudging wind farm observation data - Google Patents
Method for improving wind speed prediction of Nudging wind farm observation data Download PDFInfo
- Publication number
- CN114483485B CN114483485B CN202210170430.2A CN202210170430A CN114483485B CN 114483485 B CN114483485 B CN 114483485B CN 202210170430 A CN202210170430 A CN 202210170430A CN 114483485 B CN114483485 B CN 114483485B
- Authority
- CN
- China
- Prior art keywords
- observation data
- nudging
- wind
- forecasting
- wind speed
- 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 65
- 230000006870 function Effects 0.000 claims abstract description 20
- 239000011159 matrix material Substances 0.000 claims abstract description 18
- 238000004458 analytical method Methods 0.000 claims abstract description 14
- 230000008569 process Effects 0.000 claims abstract description 14
- 238000010801 machine learning Methods 0.000 claims abstract description 6
- 238000003908 quality control method Methods 0.000 claims abstract description 6
- 230000000007 visual effect Effects 0.000 claims abstract description 4
- 230000001419 dependent effect Effects 0.000 claims description 7
- 239000000203 mixture Substances 0.000 claims 1
- 238000013277 forecasting method Methods 0.000 abstract description 2
- 238000004422 calculation algorithm Methods 0.000 description 6
- 238000001914 filtration Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 3
- 238000012937 correction Methods 0.000 description 3
- 230000006872 improvement Effects 0.000 description 2
- 238000007637 random forest analysis Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- 238000004379 similarity theory Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000003746 surface roughness Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000010792 warming Methods 0.000 description 1
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F03—MACHINES OR ENGINES FOR LIQUIDS; WIND, SPRING, OR WEIGHT MOTORS; PRODUCING MECHANICAL POWER OR A REACTIVE PROPULSIVE THRUST, NOT OTHERWISE PROVIDED FOR
- F03D—WIND MOTORS
- F03D17/00—Monitoring or testing of wind motors, e.g. diagnostics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A30/00—Adapting or protecting infrastructure or their operation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computational Mathematics (AREA)
- Pure & Applied Mathematics (AREA)
- Computing Systems (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Sustainable Development (AREA)
- Medical Informatics (AREA)
- Evolutionary Computation (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Sustainable Energy (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- Radar Systems Or Details Thereof (AREA)
- Wind Motors (AREA)
Abstract
The invention relates to a method for improving wind speed prediction of Nudging wind farm observation data, which comprises the following steps: ⑴ Performing quality control on wind farm observation data through conventional meteorological observation data and radar observation data by adopting machine learning; ⑵ The NMC method and the EnKF method are combined, and a background error covariance matrix B is generated according to a background field of the mode forecast; ⑶ Constructing a Nudging weight function of the flow dependence; ⑷ Adopting Nudging assimilation method to assimilate wind farm observation data, conventional ground and sounding observation data, doppler weather radar and profile radar observation data continuously to obtain initial conditions; ⑸ Adding a wind power plant parameterization scheme in the WRF mode forecasting process; ⑹ Setting analysis sequence, and repeating the step ⑵~⑸ according to the mode of forecasting circulation every 3 hours; after the analysis sequence is finished, forecasting the wind speed for 2-3 days in the future; ⑺ And obtaining a visual chart and statistical data by using post-processing software. The wind power station wind speed fine forecasting method can improve the wind speed fine forecasting level of the wind power station.
Description
Technical Field
The invention relates to a wind speed forecasting method, in particular to a method for improving wind speed forecasting of Nudging wind farm observation data.
Background
Wind resources are important renewable energy sources, and wind power generation plays an important role in adjusting energy structures, reducing atmospheric pollution, relieving global warming and the like. The numerical mode is a main means of wind speed forecasting, and the wind speed forecasting directly determines the forecasting precision of wind power. Statistics show that 60% of wind power prediction errors are from numerical weather forecast.
Initial conditions, physical parameterization schemes and grid resolution are key to affecting the level of numerical weather forecast. The development of physical parameterization is slow, and the improvement of the resolution of the grid can be obtained through the nested downscaling of different scale modes, so that the construction of initial conditions is the most important, and is a preferred option for improving the mode simulation result. The multi-source observation data are reasonably assimilated by utilizing the data assimilation method, and the spatial resolution of the forecasting mode is improved by combining the improved parameterization scheme closely related to the wind power plant through multi-mode nesting, so that the method is an effective way for improving the wind speed forecasting of the wind power plant.
At present, the observation data assimilated in the domestic and foreign business forecasting systems mainly comprise: conventional weather observations (including ground observations, sounding observations, aircraft observations, ship observations, etc.), non-conventional weather observations (doppler weather radar, wind profile radar, lidar, satellites, GPS, etc.). The wind power base is generally built in areas far away from weather stations such as deserts, partition walls and grasslands, and taking China as an example, the main wind power base is located in northwest arid areas, northern grassland areas and southeast offshore areas, the available conventional and unconventional weather observation data of the wind power plant and surrounding areas are limited, and the unconventional satellite data cannot effectively acquire data on different vertical layers of the atmosphere under the condition of non-sunny days, so that the available data input of the wind power base and surrounding areas is lacking when the wind speed of the wind power plant is predicted in a numerical forecasting mode, and the improvement of the wind speed forecasting level is restricted. Although wind power bases are equipped with corresponding automatic meteorological observation systems in installed construction, for example: the wind measuring towers built by the wind power plant can realize all-weather uninterrupted observation of high time resolution of meteorological elements such as wind speed, wind direction, temperature, humidity, air pressure and the like on different height layers (30 m, 50 m, 70 m, 120 m and the like); a wind speed monitoring instrument is also arranged on each wind driven generator, so that the real-time observation of the wind speed and the wind direction at the height of the hub of the fan can be realized; in addition, part of wind power bases are additionally provided with wind profile radars according to department requirements. But the above materials have not been assimilated and incorporated into business numerical forecasting model, mainly due to: ⑴ The meteorological elements such as the wind speed of the atmosphere low layer are influenced by factors such as the terrain, the underlying surface and the like, the local benefit is very remarkable, the space representativeness of the observation data of the wind power plant is limited, and the numerical mode cannot be effectively entered; ⑵ The observation data of the wind power plant are generally not calibrated by a weather instrument regularly, the data quality needs to be further improved, and particularly, the wind power plant with years of data accumulated is subjected to the data acquisition; ⑶ The time frequency of wind power plant observation data is higher (15 min times), and under the conditions of limited calculation power and higher requirement on forecasting timeliness, certain difficulty exists in the timeliness of the high-frequency data of the same wind power plant. In conclusion, how to introduce wind speed data with small spatial scale and high time frequency observed by a wind power plant into the existing numerical forecasting mode according to an optimal assimilation way can remarkably improve the wind speed forecasting level of the numerical forecasting mode on the wind power plant.
The data assimilation method used in the domestic and foreign numerical forecasting system mainly comprises the following steps: three-dimensional, four-dimensional variations (3 DVAR, 4 DVAR), collective Kalman filtering, and newton relaxation approximation (Nudging).
The analytical field generated by the method of 3DVAR is a solution for the objective function to reach a minimum, and is usually obtained by adopting a gradual iterative minimization method. The 4DVAR method is to add a prediction mode into the observation operator of 3DVAR to realize a four-dimensional implicit covariance mode.
The Kalman filtering is performed by constructing an optimized weight matrix, calculating a weighted average of the observation field and the mode background field, and obtaining the optimal estimation of the analysis moment.
Nudging (Newton relaxation approximation) assimilation method is a mature four-dimensional assimilation method, and by constructing a weight function related to the difference between an observation field and a mode field and introducing the weight function as a forcing term into a numerical forecasting mode equation, the mode state is continuously approximated to the observation state in the mode integral forecasting process, and the forecasting level is improved. Taking a mesoscale numerical Forecasting mode WRF (WEATHER RESEARCH AND modeling) as an example, the Forecasting equation of the Nudging method is as follows:
………… （1）
The second term at the right end of the equal sign of the equation (1) is Nudging forcing term, and q is a model forecast variable such as wind speed, temperature and the like; x, y and z are three-dimensional space; t is the forecast time; f q is an explicit and implicit physical process item; n is the number of the observation data; Q 0 is the q value of the observation for Nudging weight functions with respect to space and time at a certain observation point i; q m is the q value of the pattern forecast; q 0-qm analyze delta.
The Nudging method has the advantages compared with the 3DVAR/4DVAR and collective Kalman filtering method: ⑴ The error covariance matrix is not required to be provided, the calculation process is simple, the efficiency is high, and the requirement on the calculation condition is low; ⑵ The method is a continuous data assimilation method, and observation information can be continuously introduced in the mode continuous integration process, so that mode forecast is more balanced; ⑶ The equation is simple in form, and Nudging forcing terms can be modified according to a specific physical process so as to improve the influence of local and high-frequency observation data. But also has the following disadvantages: ⑴ Only conventional observation data can be assimilated; ⑵ Weight function in Nudging equationTypically an empirical fixed value, does not contain 4DVAR or flow dependent weather evolution information in the aggregate Kalman filtering. In consideration of the characteristics of strong locality, high frequency and the like of the data observed by the wind power plant and the requirement of power dispatching on wind power prediction timeliness, the Nudging method is used for assimilating the data observed by the wind power plant, and the current-dependent weight function is introduced by increasing the influence range and the intensity of the local data, so that the method is a feasible way for realizing optimal assimilation of the wind speed data of the wind power plant.
The fan device of the wind power plant changes the surface roughness, so that the energy balance of the ground surface of the underlying surface is changed, and the important influence is generated on the atmospheric boundary layer. How to introduce the installed capacity (scale) of a wind power plant, the installation height of a fan, the size of the fan and the like into a numerical mode through a parameterization scheme, and improving the influence of the fan on an atmospheric boundary layer is a basic problem to be solved. In addition, the spatial resolution of the numerical weather forecast mode facing wind, light and electric power forecast at present is 10 km ×10 km, the occupied area of a wind power field with typical installed capacity of 5 MW is about 3 km ×3 km or even smaller, the change process of local weather is difficult to reflect in the existing numerical weather forecast, and the fine forecast level of the wind power field is required to be improved by a multi-mode nesting technology.
Disclosure of Invention
The invention aims to provide a method for improving wind speed prediction by Nudging wind farm observation data, which improves the fine prediction level of a wind farm.
In order to solve the problems, the method for improving wind speed prediction by Nudging wind farm observation data comprises the following steps:
⑴ Performing quality control on wind farm observation data with the time resolution of 15min by adopting a machine learning algorithm through conventional meteorological observation data with the time resolution of 6 h/12 h and radar observation data with the time resolution of 15 min;
⑵ By combining NMC and EnKF methods, a background error covariance matrix B is generated according to a background field of the mode forecast:
；
Wherein: b NMC is a background error covariance matrix generated by the NMC method; b EnKF is a background error covariance matrix generated by the EnKF method; b EnKF was counted during each assimilation time window.
⑶ A stream dependent Nudging weight function, i.e. a hybrid Nudging weight function W EnKF, is constructed:
Wherein: r is an observation error covariance matrix; h is an observation operator; h T is the transpose of the observation operator; superscript-1 represents matrix inversion operation; GW q is a weight function in the original Nudging equation;
⑷ Adopting Nudging assimilation method to continuously assimilate wind farm observation data with time resolution of 15 min, conventional ground and exploring observation data with time resolution of 6 h/12 h, doppler weather radar with time resolution of 15 min and profile radar observation data to obtain initial conditions;
⑸ Adding a parameterization scheme of the wind power plant in the WRF mode forecasting process;
⑹ Setting analysis sequence, and repeating the step ⑵~⑸ according to the mode of forecasting circulation every 3 hours; after the analysis sequence is finished, forecasting for 2-3 days in the future;
⑺ And (5) obtaining a visual chart and statistical data by using professional weather post-processing software.
The Nudging assimilation method in step ⑷ is performed according to the following formula:
Wherein: q is a model forecast variable; x, y and z are three-dimensional space; t is the forecast time; f q is an explicit and implicit physical process item; n is the number of the observation data; q 0 is the observed q value; q-value of q m mode forecast; q 0-qm analyze delta.
The analysis sequence in step ⑹ refers to one of 6 hours, 12 hours, or user-defined.
Compared with the prior art, the invention has the following advantages:
1. According to the invention, the quality control is carried out on the wind power plant observation data by adopting a machine learning algorithm, so that the quantity and quality of the data of the wind power plant and the surrounding areas are enriched and optimized, and the problems of insufficient data and poor data quality of the wind power plant and the surrounding areas are solved from the source of the data.
2. According to the invention, by combining an NMC method and an EnKF method, a 'flow dependent' Nudging weight function containing weather flow pattern dynamic evolution information is constructed, so that the forecast of wind speed is improved from the assimilation method.
3. According to the wind power station wind speed prediction method, a parameterization scheme that the wind power station influences the atmospheric boundary layer is considered, and the influence of the wind power station on the local atmospheric boundary layer is considered more carefully, so that the wind speed prediction can be improved.
Drawings
The following describes the embodiments of the present invention in further detail with reference to the drawings.
FIG. 1 is a schematic diagram of the present invention.
Detailed Description
As shown in FIG. 1, a method for improving wind speed prediction from Nudging wind farm observations comprises the steps of:
⑴ By adopting machine learning, the quality control of the wind farm observation data with the time resolution of 15min is carried out by the conventional meteorological observation data with the time resolution of 6 h/12 h and the radar observation data with the time resolution of 15min.
Wherein: the wind farm observation data comprise wind speed, wind direction, temperature, humidity, air pressure and the like observed on different height layers (30 m, 50m, 70 m, 120 m and the like) of a wind farm anemometer tower, and wind speed and wind direction observed at the hub height of each wind driven generator.
Generally, meteorological observation instruments at the heights of a wind measuring tower and a wind turbine hub which are deployed at the beginning of wind power plant construction are calibrated, and the reliability of the observed data is high. Therefore, basic characteristics of different heights, different positions and different observation elements of the wind power plant are obtained by combining a machine learning algorithm by using observation data of the wind power plant for 1-2 years, conventional observation data of surrounding weather stations, doppler weather radar and wind profile radar data, and a database is formed, so that local and high-frequency observation data of the wind power plant are controlled in quality. The specific process is as follows:
① Selecting weather site data (such as a conventional ground station, a sounding station, a Doppler weather radar and a wind profile radar) which are obviously related to wind farm observation data through a feature engineering (Feature engineering) learning algorithm;
② Clustering the wind power plant observation and different meteorological observations by taking the space-time continuity characteristics of different meteorological elements into consideration through a K-Means (K-Means) clustering algorithm;
③ Training to obtain a wind farm meteorological element correction model and space-time distribution characteristics of errors according to sites and observations selected in ①、② based on a long-short-term memory network (LSTM) and a Random forest (Random forest) algorithm;
④ Inputting the real-time observed wind power plant data into a ③ correction model, so as to realize the quality control of the real-time observed wind power plant data;
⑤ Aiming at the influence of the difference between the mode terrain and the altitude of the wind farm on assimilation, the following method is adopted for correction: when the difference between the altitude of the wind power plant and the mode terrain is within 100 m, the wind power plant observation is interpolated to a corresponding height above the mode terrain by utilizing a near-stratum similarity theory; otherwise, directly eliminating.
⑵ And (3) generating a background error covariance matrix B according to a background field of the mode forecast by adopting the combination of NMC and EnKF methods. B consists of a weighted average of static background error covariance (seasonal average) obtained by an NMC method and background error covariance (once every 3 hours, namely 10-20 possible states of atmosphere at the analysis time) with 'flow dependence' information generated by 10-20 aggregate disturbance samples at the analysis time. Namely:
…………………………… （2）
Wherein: b NMC is a background error covariance matrix generated by the NMC method; b EnKF is the background error covariance matrix generated by the EnKF method. B EnKF was counted during each assimilation time window.
Wherein: the background field, or first guess field, refers to the first analog/predictive value of the pattern formed by objective interpolation of the observed data or after balance control.
⑶ A stream dependent Nudging weight function, i.e. a hybrid Nudging weight function W EnKF, is constructed:
…………………………… （3）
wherein: r is an observation error covariance matrix; h is an observation operator; h T is the transpose of the observation operator; superscript-1 represents matrix inversion operation; GW q is a weight function in the original Nudging equation [ see equation (1) ], typically given empirically.
As can be seen from equations (2) - (3), since Nudging weight function W EnKF is "flow dependent," the constructed W EnKF contains dynamic evolution information of the weather patterns at the moment of analysis.
⑷ And continuously assimilating wind farm observation data with time resolution of 15 min, conventional ground and exploring observation data with time resolution of 6 h/12 h, doppler weather radar with time resolution of 15 min and profile radar observation data by adopting Nudging assimilation method to obtain initial conditions.
Wherein: substituting Nudging weight function GW q in equation (1) with the newly constructed W EnKF weight function in equation (3), equation (1) can be rewritten as:
…………… （4）
Wherein: q is a model forecast variable; x, y and z are three-dimensional space; t is the forecast time; f q is an explicit and implicit physical process item; n is the number of the observation data; q 0 is the observed q value; q-value of q m mode forecast; q 0-qm analyze delta.
Therefore, the Nudging assimilation method in the present invention is performed according to the equation (4). In order to maximize the influence of local and high-frequency observation data of the wind farm on the mode forecast result, the method is effectively integrated into a numerical mode. In the numerical mode Nudging assimilation setting process, a unidirectional feedback scheme is adopted, and the innermost layer only contains wind power plant observation data.
⑸ In order to consider the influence of the wind power plant on the local atmospheric boundary layer, a parameterization scheme of the wind power plant is added in the WRF mode forecasting process.
⑹ The analysis sequence is set, and the analysis sequence refers to one of 6 hours, 12 hours or user customization. Repeating step ⑵~⑸ in a cycle of forecasting every 6 hours; and after the analysis sequence is finished, forecasting is carried out for 2-3 days in the future.
⑺ Visual charts and statistics were obtained using professional weather post-processing software (NCL, python, etc.).
Claims (2)
1. A method of Nudging wind farm observations to improve wind speed predictions, comprising the steps of:
⑴ Performing quality control on wind speed data observed by a wind power plant with time resolution of 15 min by adopting machine learning through conventional meteorological observation data with time resolution of 6 h/12 h and radar observation data with time resolution of 15 min;
⑵ The background field of the wind speed forecasting mode is formed, and a background error covariance matrix B is generated by adopting the following method of combining NMC and EnKF methods:
Wherein: b NMC is a background error covariance matrix generated by the NMC method; b EnKF is a background error covariance matrix generated by the EnKF method; b EnKF was counted during each assimilation time window.
⑶ The construct Nudging method has a flow dependent weight function, i.e., a blend Nudging weight function W EnKF:
Wherein: r is an observation error covariance matrix; h is an observation operator; h T is the transpose of the observation operator; superscript-1 represents matrix inversion operation; GW q is a weight function in the original Nudging equation;
⑷ Adopting Nudging assimilation method to continuously assimilate wind farm observation data with time resolution of 15 min, conventional ground and exploring observation data with time resolution of 6 h/12 h, doppler weather radar with time resolution of 15 min and profile radar observation data to obtain initial conditions;
⑸ Adding a wind farm parameterization scheme in a WRF mode;
⑹ Setting analysis sequence, and repeating the step ⑵~⑸ according to the mode of forecasting circulation every 3 hours; after the analysis sequence is finished, forecasting the wind speed for 2-3 days in the future;
⑺ And (5) obtaining a visual chart and statistical data by using professional weather post-processing software.
2. A method of improving wind speed prediction from Nudging wind farm observations as claimed in claim 1, wherein: the Nudging assimilation method in step ⑷ is performed according to the following formula:
Wherein: q is a model forecast variable; x, y and z are three-dimensional space; t is time; f q is an explicit and implicit physical process item; n is the number of the observation data; q 0 is the observed value of q; q m is the q value of the pattern forecast; q 0-qm analyze the field increment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210170430.2A CN114483485B (en) | 2022-02-24 | 2022-02-24 | Method for improving wind speed prediction of Nudging wind farm observation data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210170430.2A CN114483485B (en) | 2022-02-24 | 2022-02-24 | Method for improving wind speed prediction of Nudging wind farm observation data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114483485A CN114483485A (en) | 2022-05-13 |
| CN114483485B true CN114483485B (en) | 2024-07-16 |
Family
ID=81483588
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210170430.2A Active CN114483485B (en) | 2022-02-24 | 2022-02-24 | Method for improving wind speed prediction of Nudging wind farm observation data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114483485B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116304491B (en) * | 2023-05-11 | 2023-08-08 | 长江三峡集团实业发展(北京)有限公司 | Assimilation method and system for marine anomaly observation data |
| CN117574748B (en) * | 2023-10-08 | 2024-05-10 | 兰州大学 | Plateau vortex initialization method with balanced power and hydrothermal structure |
| CN118100165B (en) * | 2024-03-05 | 2024-07-26 | 兰州大学 | Method for improving wind power prediction by using radar wind measurement data of homologous engine room |
| CN118349956A (en) * | 2024-04-29 | 2024-07-16 | 粤港澳大湾区气象监测预警预报中心(深圳气象创新研究院) | Weather management engineering method, system, equipment and storage medium |
| CN118228204B (en) * | 2024-05-22 | 2024-10-29 | 南京智汇环境气象产业研究院有限公司 | Emission source inversion method combining extreme stochastic tree and Newton relaxation approximation |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112149859A (en) * | 2019-06-28 | 2020-12-29 | 北京金风慧能技术有限公司 | Wind speed prediction method and wind speed prediction device |
| CN113236507A (en) * | 2021-05-28 | 2021-08-10 | 中南大学 | Yaw static error diagnosis method and system for wind turbine generator |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11113395B2 (en) * | 2018-05-24 | 2021-09-07 | General Electric Company | System and method for anomaly and cyber-threat detection in a wind turbine |
-
2022
- 2022-02-24 CN CN202210170430.2A patent/CN114483485B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112149859A (en) * | 2019-06-28 | 2020-12-29 | 北京金风慧能技术有限公司 | Wind speed prediction method and wind speed prediction device |
| CN113236507A (en) * | 2021-05-28 | 2021-08-10 | 中南大学 | Yaw static error diagnosis method and system for wind turbine generator |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114483485A (en) | 2022-05-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114483485B (en) | Method for improving wind speed prediction of Nudging wind farm observation data | |
| TWI476430B (en) | A wind energy forecasting method with extreme wind speed prediction function | |
| Bentzien et al. | Generating and calibrating probabilistic quantitative precipitation forecasts from the high-resolution NWP model COSMO-DE | |
| CN109919353A (en) | A kind of distributed photovoltaic prediction technique of the ARIMA model based on spatial coherence | |
| Di et al. | Improving WRF model turbine-height wind-speed forecasting using a surrogate-based automatic optimization method | |
| CN109447315A (en) | A kind of electric power meteorology numerical weather forecast method and apparatus based on multiple space and time scales | |
| Mabel et al. | Estimation of energy yield from wind farms using artificial neural networks | |
| CN101741085A (en) | Method for forecasting short-term wind-electricity power | |
| CN102945508A (en) | Model correction based wind power forecasting system and method | |
| CN103258242A (en) | Wind measurement network layout method based on wind power plant layout in large-scale wind power base | |
| CN112541654B (en) | Regional wind energy resource refined assessment method | |
| CN105279576A (en) | Wind speed forecasting method | |
| Chen et al. | Improving afternoon thunderstorm prediction over Taiwan through 3DVar-based radar and surface data assimilation | |
| Karamanis et al. | Wind energy resources in the Ionian Sea | |
| CN116738769B (en) | Wind-light data generation model construction method and wind-light data generation method | |
| CN113075751A (en) | Method and system for fusing observation data in short-term forecasting | |
| CN112541655A (en) | Atmospheric re-analysis method for refined assessment demand of regional wind energy resources | |
| Poitras et al. | Wind speed prediction for a target station using neural networks and particle swarm optimization | |
| CN115841265A (en) | High-resolution onshore wind power generation resource evaluation method and system | |
| CN112801429B (en) | Micro-terrain wind power calculation method, device and system | |
| CN117634701B (en) | Solar radiation short-term forecasting method and system based on regional photovoltaic set forecasting | |
| Koung et al. | Technoeconomic Analysis of a Wind Power Generation System and Wind Resource Mapping Using GIS Tools: The Case of Twelve Locations in the Commune of Evodoula, Cameroon | |
| CN118100165B (en) | Method for improving wind power prediction by using radar wind measurement data of homologous engine room | |
| Qu et al. | Temperature Interpolation Method of Distributed Photovoltaic Power Station Group Based on Multisource Data Fusion | |
| Jiang | A hybrid times-scale assessment method for wind power density in complex mountainous areas based on improved Xgboost |
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 |