CN117175585B - Wind power prediction method, device, equipment and storage medium - Google Patents
Wind power prediction method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN117175585B CN117175585B CN202311445250.1A CN202311445250A CN117175585B CN 117175585 B CN117175585 B CN 117175585B CN 202311445250 A CN202311445250 A CN 202311445250A CN 117175585 B CN117175585 B CN 117175585B
- Authority
- CN
- China
- Prior art keywords
- wind power
- data
- vortex
- predicted
- learning model
- 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 77
- 238000007781 pre-processing Methods 0.000 claims abstract description 24
- 238000012549 training Methods 0.000 claims description 60
- 238000005259 measurement Methods 0.000 claims description 45
- 238000012360 testing method Methods 0.000 claims description 29
- 238000004458 analytical method Methods 0.000 claims description 20
- 239000002245 particle Substances 0.000 claims description 19
- 230000008859 change Effects 0.000 claims description 18
- 238000013528 artificial neural network Methods 0.000 claims description 15
- 238000011156 evaluation Methods 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 12
- 238000012706 support-vector machine Methods 0.000 claims description 12
- 238000010276 construction Methods 0.000 claims description 5
- UHZZMRAGKVHANO-UHFFFAOYSA-M chlormequat chloride Chemical compound [Cl-].C[N+](C)(C)CCCl UHZZMRAGKVHANO-UHFFFAOYSA-M 0.000 claims 2
- 238000010248 power generation Methods 0.000 abstract description 8
- 230000008569 process Effects 0.000 description 10
- 230000006870 function Effects 0.000 description 9
- 238000004891 communication Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000013507 mapping Methods 0.000 description 4
- 238000012795 verification Methods 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 230000007774 longterm Effects 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 description 1
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 description 1
- 238000005311 autocorrelation function Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000002203 pretreatment Methods 0.000 description 1
- 238000007670 refining Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000007704 transition Effects 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
- Wind Motors (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the field of wind power generation and discloses a wind power prediction method, a device, equipment and a storage medium, wherein the method obtains standard data to be predicted by obtaining the data to be predicted of a wind power plant and performing data preprocessing on the data to be predicted; determining vortex characteristics of the wind power plant according to standard data to be predicted; and predicting the standard data to be predicted and the vortex characteristics through a hybrid learning model to obtain the wind power of the wind power plant. The vortex characteristics of the wind power plant are determined according to the standard data to be predicted of the wind power plant, and the vortex characteristics and the standard data to be predicted are predicted through the hybrid learning model, so that the data dimension of wind power prediction is increased, and the accuracy of wind power prediction is improved.
Description
Technical Field
The invention relates to the technical field of wind power generation, in particular to a wind power prediction method, a device, equipment and a storage medium.
Background
The output power of wind power depends on meteorological factors such as wind speed, temperature, wind direction, pressure and the like, but the stability of a power grid is easy to be impacted due to uncertainty and intermittence of the meteorological factors. Therefore, prediction of wind power is required to improve the stability of the grid.
The traditional wind power prediction method is mainly based on a physical model, such as a wind speed coupled energy balance model, a time sequence model and the like. The method needs to have high requirements on aerodynamics of the wind turbine, and is difficult to accurately model, so that prediction accuracy is low.
The foregoing is provided merely for the purpose of facilitating understanding of the technical solutions of the present invention and is not intended to represent an admission that the foregoing is prior art.
Disclosure of Invention
The invention mainly aims to provide a wind power prediction method, a device, equipment and a storage medium, and aims to solve the technical problems that the traditional wind power prediction method in the prior art has higher requirements on aerodynamics of a wind turbine generator set, is difficult to accurately model, and causes lower prediction accuracy.
In order to achieve the above object, the present invention provides a wind power prediction method, which includes the steps of:
obtaining data to be predicted of a wind power plant, and carrying out data preprocessing on the data to be predicted to obtain standard data to be predicted;
determining vortex characteristics of the wind farm according to the standard data to be predicted;
and predicting the standard data to be predicted and the vortex characteristics through a hybrid learning model to obtain the wind power of the wind power plant.
Optionally, the step of determining the vortex characteristics of the wind farm according to the standard data to be predicted includes:
processing the standard data to be predicted through a particle swarm algorithm to determine the vortex center position;
acquiring a wind field change condition within a preset range of the vortex center position, and optimizing the vortex center position according to the wind field change condition to obtain a final vortex center position;
and determining vortex characteristics of each final vortex center position in the wind power plant according to the final vortex center position and the standard data to be predicted.
Optionally, the step of predicting the standard data to be predicted and the vortex characteristics through a hybrid learning model to obtain wind power of the wind farm includes:
predicting the standard data to be predicted and the vortex characteristics at the final vortex center position through a hybrid learning model to obtain wind power of each wind turbine in the wind power plant;
and determining the wind power of the wind power plant according to the wind power of each wind turbine generator.
Optionally, before the step of obtaining the data to be predicted of the wind farm and performing data preprocessing on the data to be predicted, the method further includes:
Acquiring historical measurement data of a wind power plant, and preprocessing the historical measurement data to obtain standard historical data;
performing vortex characteristic analysis on the standard historical data to determine vortex characteristics corresponding to the standard historical data set;
generating a training data set and a test data set according to the standard historical data and the vortex characteristics;
inputting the training data set into an initial hybrid learning model for training to obtain a premixed learning model;
performing precision evaluation on the pre-mixing learning model through a test data set to obtain model precision of the pre-mixing learning model;
and generating a hybrid learning model when the model precision of the pre-mixed learning model is higher than the preset precision.
Optionally, the step of inputting the training data set into an initial hybrid learning model for training to obtain a premixed learning model includes:
constructing an initial hybrid learning model based on a BP neural network and a support vector machine;
and updating and training the BP neural network and the support vector machine through the training data set to obtain a pre-mixing learning model.
Optionally, before the step of generating the training data set and the test data set according to the standard history data and the vortex characteristics, the method further comprises:
Acquiring a start-stop state of a wind turbine in a wind power plant, and generating start-stop parameters of the wind turbine according to the start-stop state of the wind turbine;
accordingly, the step of generating a training data set and a test data set from the standard history data and the vortex characteristics comprises:
and generating a training data set and a testing data set according to the standard historical data, the wind turbine generator start-stop parameters and the vortex characteristics.
Optionally, after the step of obtaining the model accuracy of the pre-mixing learning model by evaluating the accuracy of the pre-mixing learning model by using a test data set, the method further includes:
and when the model precision of the pre-mixing learning model is not higher than the preset precision, returning to the step of acquiring the historical measurement data of the wind power plant, preprocessing the historical measurement data and obtaining the standard historical data until the model precision of the pre-mixing learning model is higher than the preset precision.
In addition, in order to achieve the above object, the present invention also provides a wind power prediction apparatus, including:
the data processing module is used for acquiring data to be predicted of the wind power plant, and carrying out data preprocessing on the data to be predicted to obtain standard data to be predicted;
The vortex analysis module is used for determining the vortex characteristics of the wind power plant according to the standard data to be predicted;
and the power prediction module is used for predicting the standard data to be predicted and the vortex characteristics through a hybrid learning model to obtain the wind power of the wind power plant.
In addition, in order to achieve the above object, the present invention also proposes a wind power prediction apparatus, the apparatus comprising: a memory, a processor, and a wind power prediction program stored on the memory and executable on the processor, the wind power prediction program configured to implement the steps of the wind power prediction method as described above.
In addition, in order to achieve the above object, the present invention also proposes a storage medium having stored thereon a wind power prediction program which, when executed by a processor, implements the steps of the wind power prediction method as described above.
According to the method, standard data to be predicted are obtained by acquiring the data to be predicted of the wind power plant and carrying out data preprocessing on the data to be predicted; determining vortex characteristics of the wind power plant according to standard data to be predicted; and predicting the standard data to be predicted and the vortex characteristics through a hybrid learning model to obtain the wind power of the wind power plant. The vortex characteristics of the wind power plant are determined according to the standard data to be predicted of the wind power plant, and the vortex characteristics and the standard data to be predicted are predicted through the hybrid learning model, so that the data dimension of wind power prediction is increased, and the accuracy of wind power prediction is improved.
Drawings
FIG. 1 is a schematic structural diagram of a wind power prediction device of a hardware operating environment according to an embodiment of the present invention;
FIG. 2 is a flowchart of a wind power prediction method according to a first embodiment of the present invention;
FIG. 3 is a flowchart of a second embodiment of a wind power prediction method according to the present invention;
FIG. 4 is a flowchart of a third embodiment of a wind power prediction method according to the present invention;
FIG. 5 is a flowchart of an application scenario of the wind power prediction method of the present invention;
FIG. 6 is a block diagram of a wind power prediction apparatus according to a first embodiment of the present invention.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Referring to fig. 1, fig. 1 is a schematic structural diagram of a wind power prediction device in a hardware operation environment according to an embodiment of the present invention.
As shown in fig. 1, the wind power prediction apparatus may include: a processor 1001, such as a central processing unit (Central Processing Unit, CPU), a communication bus 1002, a user interface 1003, a network interface 1004, a memory 1005. Wherein the communication bus 1002 is used to enable connected communication between these components. The user interface 1003 may include a Display, an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may further include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a Wireless interface (e.g., a Wireless-Fidelity (Wi-Fi) interface). The Memory 1005 may be a high-speed random access Memory (Random Access Memory, RAM) or a stable nonvolatile Memory (NVM), such as a disk Memory. The memory 1005 may also optionally be a storage device separate from the processor 1001 described above.
Those skilled in the art will appreciate that the structure shown in FIG. 1 is not limiting of a wind power plant and may include more or fewer components than shown, or may combine certain components, or a different arrangement of components.
As shown in fig. 1, an operating system, a network communication module, a user interface module, and a wind power prediction program may be included in the memory 1005 as one type of storage medium.
In the wind power prediction apparatus shown in fig. 1, the network interface 1004 is mainly used for data communication with a network server; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the wind power prediction device of the present invention may be disposed in the wind power prediction device, where the wind power prediction device invokes a wind power prediction program stored in the memory 1005 through the processor 1001, and executes the wind power prediction method provided by the embodiment of the present invention.
The embodiment of the invention provides a wind power prediction method, and referring to fig. 2, fig. 2 is a schematic flow chart of a first embodiment of the wind power prediction method.
It should be noted that, in order to solve the problem that in the prior art, the traditional wind power prediction method is mainly based on a physical model, has higher requirements on aerodynamics of a wind turbine generator, and is difficult to accurately model, so that the prediction precision is lower. According to the invention, a mixed learning model is established by adopting a mixed learning algorithm combining the BP neural network and the support vector machine (Support Vector Machine, SVM), the advantages of the two methods are comprehensively considered, the robustness of power prediction is improved, and the prediction precision is further improved. According to the method, meteorological factors and the influence of a single wind turbine on the power of the wind power plant are considered, meteorological data such as temperature, humidity and pressure, the power of the wind turbine, vortex characteristics of the wind turbine, start-stop characteristics of the wind turbine and the like are taken as the input of the hybrid learning model, so that the power of the wind power plant is predicted from multiple dimensions, the power is refined to the single machine prediction of the wind turbine, and the distributed control of the wind power plant is realized.
It will be appreciated that a hybrid learning model needs to be built before the wind power prediction method is performed. The method for establishing the hybrid learning model can be as follows:
step S01: and acquiring historical measurement data of the wind power plant, and preprocessing the historical measurement data to obtain standard historical data.
It should be understood that the device for performing hybrid learning modeling may be the wind power prediction device described above, or may be another device with a data modeling function, which is not limited in this embodiment. This embodiment and the following embodiments will be described in detail with reference to wind power prediction devices as examples.
The method is characterized in that the input and output variables of a hybrid learning algorithm are formed by acquiring historical measurement data such as wind speed, wind direction, temperature, pressure and wind turbine generator power in the past N hours of a wind farm, so that the construction of a hybrid learning model is realized.
It can be understood that when the historical measurement data of the wind farm is obtained, a preset data preprocessing method can be adopted to detect abnormal values and outliers in the data, so that the historical measurement data is subjected to smoothing processing, noise data in the historical measurement data are filtered, and the subsequent modeling effect is improved.
It should be understood that in this embodiment, in consideration of the influence of meteorological factors on wind power, meteorological data such as temperature, humidity, pressure and the like are used as input variables of an initial hybrid learning model, so that prediction accuracy under certain meteorological conditions is further improved.
Specifically, the historical measurement data may be smoothed by detecting the LOF values for each data point in the historical measurement data and identifying outliers in the historical measurement data using an LOF algorithm.
The smoothing method may also include a Z-score method based on a mean value and a standard deviation, a box-plot method, and the like, which is not limited in this embodiment.
In specific implementation, the wind power prediction equipment acquires historical measurement data of a wind power plant, removes abnormal values and outliers in the historical measurement data, and acquires standard historical data.
Step S02: and carrying out vortex characteristic analysis on the standard historical data, and determining vortex characteristics corresponding to the standard historical data set.
By carrying out vortex characteristic analysis on the standard historical data, the central position of each vortex in the wind power plant and the vortex characteristics corresponding to the vortex can be determined.
Compared with the prior art, the hybrid learning model provided by the invention takes the vortex center position of the vortex in the wind power plant and the vortex characteristics corresponding to the vortex as additional inputs, so that the hybrid learning algorithm can be helped to establish a more accurate mapping relation, and the prediction precision of the hybrid learning model is improved. In addition, a learning model for predicting wind power of a wind power plant can be accurately established without the requirement on aerodynamic design of the wind turbine generator.
In order to perform vortex characteristic analysis on standard historical data, first, a vortex center position of each vortex in a wind power plant is determined. The method for determining the center position of the vortex may be a vortex detection algorithm, an autocorrelation function analysis method, or the like, which is not limited in this embodiment.
It will be appreciated that further analysis may be performed on the wind speed and direction data around the vortex center in determining the vortex center position to extract vortex characteristics, such as wind power variation, vortex editing, vortex strength, periodicity, etc. characteristics of the vortex.
It should be appreciated that the extracted vortex features may be used as additional input features to the hybrid learning model to further enhance the prediction accuracy of the hybrid learning model.
In the specific implementation, the wind power prediction device performs vortex characteristic analysis on standard historical data, determines vortex center positions corresponding to each vortex in the wind power plant and vortex characteristics corresponding to the vortex, and takes the vortex characteristics as additional input characteristics of the hybrid learning model so as to further enhance the prediction accuracy of the hybrid learning model.
Step S03: and generating a training data set and a testing data set according to the standard historical data and the vortex characteristics.
It will be appreciated that to obtain training data for the hybrid learning model, standard historical data and vortex characteristics may be divided into training data sets and test data sets. The specific dividing ratio may be that 80% is a training data set and 20% is a test data set; or 70% is a training data set, 30% is a test data set, and the embodiment does not limit the division ratio of the training data, and can divide according to the requirements in practical application.
It should be appreciated that the training data set, i.e. the data used to train the initial hybrid learning model, by training the initial hybrid learning model, a pre-mixed learning model can be obtained that can be used to predict wind farm data. Specifically, step S04: and inputting the training data set into an initial hybrid learning model for training to obtain a premixed learning model.
It should be explained that, in this embodiment, a hybrid learning algorithm based on a BP neural network and a support vector machine is used to construct an initial hybrid learning model, and a training data set is used as input data of the initial hybrid learning model, so as to implement training of the initial hybrid learning model and obtain a pre-training model.
It can be understood that the BP neural network can fit a complex nonlinear relation, and the support vector machine has stronger generalization capability. By combining the BP neural network and the support vector machine to construct an initial hybrid learning model, the advantages of the two parties can be combined, the prediction accuracy of wind power is improved, and the prediction robustness is improved.
Specifically, the node numbers of an input layer, a hidden layer and an output layer of the BP neural network can be set, and a weight value is initialized; a data sample is randomly selected from the training data set to be used as the input of the BP neural network, the characteristics of the data sample are input into the network, and the output result of the data sample is calculated through forward propagation.
At the same time, a suitable function is selected as a kernel function of the SVM model, such as a Gaussian kernel function, a linear kernel function, etc. And corresponding parameters, such as penalty coefficients, kernel function parameters, and the like, are selected for the SVM model. And training the SVM model according to the training data set to obtain the SVM model capable of wind power prediction.
Further, the BP neural network and the SVM model are fused, and a pre-neural network model can be obtained. The mode of performing model fusion may be weighted fusion, specifically may be setting a model weight of a BP neural network and a model weight of an SVM model, and performing weighted calculation on an output result according to the respective model weights, thereby obtaining a fused output result.
It should be understood that the model weight of each model may be determined according to factors such as model performance and model confidence in practical applications, which is not limited in this embodiment.
In order to further improve the prediction accuracy of the hybrid learning model, the hybrid learning model may be further optimized by a test data set.
Specifically, step S05: and carrying out precision evaluation on the pre-mixing learning model through a test data set to obtain the model precision of the pre-mixing learning model.
Step S06: generating a hybrid learning model when the model precision of the pre-mixed learning model is higher than a preset precision;
it can be understood that the accuracy evaluation is performed on the pre-mixing learning model, and whether the pre-mixing learning model is qualified or not is judged according to the accuracy evaluation result. If the pre-mixing learning model is qualified, the pre-mixing learning model can be used as a mixed learning model to predict the wind power of the wind power plant. If the pre-mixed learning model is unqualified, the training data needs to be updated, and the pre-mixed learning model is re-established according to the updated training data. That is, after the step of obtaining the model accuracy of the pre-mixing learning model by evaluating the accuracy of the pre-mixing learning model by the test data set, the method further includes:
And when the model precision of the pre-mixing learning model is not higher than the preset precision, returning to the step of acquiring the historical measurement data of the wind power plant, preprocessing the historical measurement data and obtaining the standard historical data until the model precision of the pre-mixing learning model is higher than the preset precision.
The accuracy evaluation may be performed by using verification indexes such as mean absolute error (mean absolute error, MAE), root mean square error (Root Mean Squared Error, RMSE), mean absolute percentage error (Mean absolute percentage error, MAPE), and the like. In the embodiment, the collected historical measurement data is divided into a training data set and a test data set by adopting a set aside method, and the average value of verification indexes is calculated through multiple times of cross verification, so that the reliability of precision evaluation is improved, and a verification index threshold value is set to judge whether the performance of the pre-mixed learning model can meet the requirement.
It can be understood that the accuracy evaluation is performed on the pre-mixing learning model, so that the generated mixed learning model can meet the preset model accuracy requirement, and the prediction accuracy of wind power is improved.
It should be appreciated that after the hybrid learning model is obtained, the wind power of the wind farm may be predicted by the hybrid learning model.
In this embodiment, the wind power prediction method includes the following steps:
step S10: and obtaining data to be predicted of the wind power plant, and carrying out data preprocessing on the data to be predicted to obtain standard data to be predicted.
It should be noted that, the execution body of the method in this embodiment may be a terminal device having functions of wind power prediction, data processing, and program running, for example, a computer, a server, etc., or may be an electronic device having the same or similar functions, for example, the wind power prediction device described above. Hereinafter, this embodiment and the following embodiments will be described with reference to a wind power prediction apparatus (hereinafter, referred to as a prediction apparatus).
It can be appreciated that the wind farm is an area where wind power generation is performed, and a plurality of wind turbines may be disposed in the wind farm. Wind power generation of a wind power plant can be realized through the wind turbine generator.
It should be understood that the data to be predicted is wind power parameters that can predict the power of the wind power plant, such as wind direction, temperature, humidity, pressure, wind speed and other wind power plant measurement data. The data to be predicted may be real-time data or non-real-time data, which is not limited in this embodiment.
It will be appreciated that a digital anemometer or other wind speed measurement device may be mounted on the wind turbine, through which the wind speed of the wind turbine when wind power is being generated may be measured to determine wind speed data.
It should be appreciated that other wind farm measurement data may be measured by the same or similar measurement devices as wind speed, which is not limited in this embodiment.
When wind farm measurement data is obtained, the obtained measurement data can be preprocessed, so that the data quality of the wind farm measurement data is improved, and power prediction is better performed. The pretreatment method can comprise the following steps: the processing modes such as filling of missing values, detection and repair of abnormal values, and standardization are not limited in this embodiment.
It can be understood that the data to be predicted is preprocessed, so that standard data to be predicted with higher data quality can be obtained, redundant data interference in detected measurement data is avoided, and the accuracy of power prediction is improved.
In the specific implementation, the prediction equipment obtains standard data to be predicted with higher data quality by obtaining the data to be predicted of the wind power plant measured by the measurement equipment and carrying out data preprocessing on the data to be predicted.
Step S20: and determining vortex characteristics of the wind farm according to the standard data to be predicted.
The vortex characteristics of the wind power plant, namely the vortex characteristics of each wind turbine generator set in the wind power plant, are described. The swirl feature may include a swirl center position, a swirl radius, a swirl direction, a swirl strength, periodicity, and the like. According to the method, the vortex characteristics of each wind turbine are used as additional characteristics of judging wind turbines, the power of a wind power plant can be predicted more accurately, the method is suitable for performance change of the wind turbines in a long-term running process, and prediction stability is guaranteed while prediction accuracy is achieved.
It can be understood that by taking the vortex characteristics of each vortex in the wind power plant as the additional input characteristics of the hybrid learning model, the influence of the vortex on the aspects of power output, fatigue load and the like of the wind turbine can be better understood, and the wind power prediction accuracy is improved.
In specific implementation, the prediction equipment performs vortex characteristic analysis according to standard data to be predicted, determines vortex characteristics of the wind power plant, takes the vortex characteristics as additional input characteristics of the hybrid learning model, and further improves prediction accuracy and stability of the hybrid learning model for wind power prediction.
Step S30: and predicting the standard data to be predicted and the vortex characteristics through a hybrid learning model to obtain the wind power of the wind power plant.
It can be understood that when the standard data to be predicted and the vortex characteristics of the vortex in the wind power plant are obtained, the standard data to be predicted and the vortex characteristics can be predicted through the hybrid learning model, and the wind power of the wind power plant is obtained. The wind power prediction of the wind power plant is carried out according to the input data to be predicted, and the prediction is carried out through a mixed learning model based on the BP neural network and the support vector machine, so that the prediction precision is improved; the vortex characteristic of vortex in the wind power plant is used as an additional input characteristic, so that the power prediction of the wind power plant is refined to the motor prediction of each wind turbine, the distributed wind power plant power prediction is realized, and the prediction accuracy is further improved.
In the specific implementation, the prediction equipment predicts standard data to be predicted and vortex characteristics through a hybrid learning model, so that the wind power of the wind power plant is obtained, and the wind power prediction accuracy is improved.
According to the embodiment, standard data to be predicted is obtained by acquiring the data to be predicted of the wind power plant and carrying out data preprocessing on the data to be predicted; determining vortex characteristics of the wind power plant according to standard data to be predicted; and predicting the standard data to be predicted and the vortex characteristics through a hybrid learning model to obtain the wind power of the wind power plant. The vortex characteristics of the wind power plant are determined according to the standard data to be predicted of the wind power plant, and the vortex characteristics and the standard data to be predicted are predicted through the hybrid learning model, so that the data dimension of wind power prediction is increased, and the accuracy of wind power prediction is improved.
Referring to fig. 3, fig. 3 is a flowchart illustrating a wind power prediction method according to a second embodiment of the present invention.
Based on the first embodiment, in this embodiment, in order to perform vortex characteristic analysis on standard data to be predicted, the step of determining vortex characteristics of the wind farm according to the standard data to be predicted includes:
step S21: processing the standard data to be predicted through a particle swarm algorithm to determine the vortex center position;
step S22: and acquiring the wind field change condition within the preset range of the vortex center position, and optimizing the vortex center position according to the wind field change condition to obtain the final vortex center position.
It should be explained that, in this embodiment, when the standard data to be predicted is obtained, the rotational angular velocity may be calculated according to the wind direction data and the wind speed data in the standard data to be predicted. And calculating wind field change conditions in a preset range of each vortex center position by utilizing a particle swarm algorithm according to the rotation angular velocity and the standard data area-level vortex center position to be predicted, so as to determine the final vortex center position.
In this embodiment, the standard data to be predicted is processed through a particle swarm algorithm, so as to determine the vortex center position in the wind power plant.
Specifically, an objective function may be defined that may measure the confidence of the vortex center position based on the degree of difference in the standard to-be-predicted data such as wind speed and wind direction. The smaller the difference, the more consistent the vortex center position and the wind speed and wind direction values in the preset range can be indicated.
By setting the initial position and velocity of the particle swarm and assigning a random initial position to each particle. By representing the position of each particle as an estimate of the vortex center, the vortex center position can be determined from the particle position. For each particle's position, its fitness can be evaluated by calculating the degree of difference in wind direction and wind speed within a preset range around it.
It should be noted that the preset range may be 3m, 4m, 5m or other distance ranges around the particle, which is not limited in this embodiment. The index for measuring the difference degree may be a euclidean distance, a cosine similarity, or other indexes, which is not limited in this embodiment.
It should be appreciated that the wind field variation may include wind speed variation, wind direction variation, temperature variation, pressure variation, etc.
It can be understood that according to the updating rule of the particle swarm algorithm, the position and the speed of each particle can be updated, and the optimal vortex center position, namely the final vortex center position, can be obtained by continuously and iteratively adjusting the positions of the particles.
It should be noted that, in model training, a particle swarm algorithm may be used as the vortex analysis method of the present invention, which is not limited in this embodiment.
In specific implementation, the prediction equipment processes standard data to be predicted through a particle swarm algorithm, and the vortex center position of vortex in the wind power plant is determined; and optimizing the vortex center position according to the wind field change condition within the preset range of the vortex center position, and determining the final vortex center position.
Step S23: and determining vortex characteristics of each final vortex center position in the wind power plant according to the final vortex center position and the standard data to be predicted.
It is understood that when determining the final vortex center position of the wind power plant, the vortex characteristics corresponding to each vortex, such as the vortex radius, wind direction change, wind speed gradient, periodicity, vortex strength, angular velocity and the like, can be determined according to the final vortex center position of each vortex.
Correspondingly, the step of predicting the standard data to be predicted and the vortex characteristics through a hybrid learning model to obtain the wind power of the wind power plant comprises the following steps:
step S31: and predicting the standard data to be predicted and the vortex characteristics at the final vortex center position through a hybrid learning model to obtain the wind power of each wind turbine in the wind power plant.
Step S32: and determining the wind power of the wind power plant according to the wind power of each wind turbine generator.
It can be understood that when the standard band prediction data and the vortex characteristics of the final vortex center position are obtained, the wind power of each wind turbine in the wind power plant can be determined according to the standard to-be-predicted data and the vortex characteristics, and then the prediction power of each wind turbine is summarized, so that the prediction power of the whole wind power plant can be obtained. Therefore, distributed wind power plant control is realized, the robustness of the hybrid learning model is improved, the nonlinear mapping relation of the wind turbine generator is more accurately established, and high-precision prediction is realized. Meanwhile, the method can adapt to performance change in the long-term running process of the wind turbine, and ensure prediction stability while realizing prediction accuracy.
In the embodiment, standard data to be predicted are processed through a particle swarm algorithm, and the vortex center position is determined; acquiring a wind field change condition within a preset range of the vortex center position, and optimizing the vortex center position according to the wind field change condition to obtain a final vortex center position; and determining the vortex characteristics of each final vortex center position in the wind power plant according to the final vortex center position and standard data to be predicted. The vortex characteristics of the vortex in the wind power plant are determined according to the final vortex center position, so that the power prediction of the wind power plant is refined to single machine prediction, the prediction precision is improved, and the distributed wind power plant control is realized.
Based on the above embodiments, in order to further improve the accuracy of wind power prediction, a third embodiment of the method of the present invention is proposed, and referring to fig. 4, fig. 4 is a schematic flow chart of the third embodiment of the wind power prediction method of the present invention.
In this embodiment, before the step of generating the training data set and the test data set according to the standard history data and the vortex characteristics, the method further includes:
step S02A: acquiring a start-stop state of a wind turbine in a wind power plant, and generating start-stop parameters of the wind turbine according to the start-stop state of the wind turbine;
accordingly, the step of generating a training data set and a test data set from the standard history data and the vortex characteristics comprises:
step S03A: and generating a training data set and a testing data set according to the standard historical data, the wind turbine generator start-stop parameters and the vortex characteristics.
It should be noted that, considering that the power of the wind turbine generator changes drastically during the start-up and stop, the prediction difficulty in this period is relatively high. To prevent power variations during start-up and shut-down from interfering with the prediction, the present embodiment introduces start-stop parameters for representing the start-stop state of the wind turbine. And the start-stop parameters are used as additional input parameters of the initial hybrid learning model of the wind turbine generator to train the initial hybrid learning model, so that the prediction accuracy of the wind turbine generator in the transition period in the starting and stopping processes is improved, and the accuracy of wind power prediction is further improved.
It can be understood that by introducing the start-stop parameters, when the hybrid learning model is constructed, the start-stop parameters need to be added into the corresponding training data set and test data set to realize training and prediction of the initial hybrid learning model, and a hybrid learning model for predicting wind power of the wind power plant is generated.
It should be appreciated that when the hybrid learning model is applied, start-stop parameters of the wind turbine can be obtained as well, so that wind power of the wind power plant can be predicted more accurately.
According to the embodiment, the starting and stopping states of the wind turbine in the wind power plant are obtained, and the starting and stopping parameters of the wind turbine are generated according to the starting and stopping states of the wind turbine; and training the initial hybrid learning model by taking the start-stop parameters as additional input features. By considering the situation that the power of the wind turbine generator is changed severely during starting and stopping, the starting and stopping parameters are introduced to conduct special prediction, and the prediction accuracy of the wind turbine generator during starting and stopping is improved.
As shown in fig. 5, fig. 5 is a flowchart of an application scenario of the wind power prediction method of the present invention.
Referring to fig. 5, in one implementation, standard historical measurement data is obtained by acquiring historical measurement data and performing a data preprocessing operation on the historical measurement data. By performing a vortex signature analysis on the standard historical measurement data, the vortex signature of the vortex in the wind farm can be further determined. Meanwhile, start-stop parameters can be introduced to specially predict the power during the start and stop of the wind turbine generator.
Further, a training data set is determined according to standard historical measurement data, vortex characteristics and start-stop parameters, and a hybrid learning model is built through a hybrid learning algorithm. Training the hybrid learning model through the training data set, carrying out precision evaluation on the output of the hybrid learning model, and judging the model precision of the hybrid learning model according to the precision evaluation result.
Further, whether the required model precision is reached is judged according to the model precision. If the accuracy is not reached, the new history measurement data and the training data set corresponding to the new history measurement data need to be re-acquired, and the step of constructing the hybrid learning model through the hybrid learning algorithm is returned to, so that the training data set corresponding to the new history measurement data is predicted, and the hybrid learning model is updated. By iterative updating multiple times until the hybrid learning model reaches accuracy, a hybrid learning model that can be used for power prediction is obtained.
Further, by acquiring data such as real-time wind speed, wind direction, temperature and pressure of the wind power plant as input, calculating vortex characteristics in real time, inputting the vortex characteristics into a hybrid learning model for prediction, realizing real-time prediction of wind power generation unit power, and summarizing the predicted power of each wind power generation unit to obtain the predicted power of the whole wind power plant.
According to the wind power generation set nonlinear mapping relation prediction method, the hybrid learning model is established by adopting the hybrid learning algorithm, vortex characteristics of vortex in the wind power plant are combined, nonlinear mapping relation of the wind power generation set can be established more accurately, and high-precision prediction of wind power of the wind power plant is realized. The method can update the learning algorithm by acquiring the real-time wind farm data on line, can be the performance change of the English wind turbine during the long-term operation process, and ensures the prediction stability while realizing the prediction precision.
Meanwhile, the invention adopts a hybrid learning algorithm and a data analysis method, is easy for engineering realization, has lower technical complexity and better feasibility. Meanwhile, the hybrid learning algorithm has higher real-time performance, and can rapidly complete prediction under lower calculated amount, thereby meeting the application requirements of wind power forecasting.
In addition, the method can predict the power of individual fans by refining to single-machine prediction, provides a basis for distributed control and correction of the wind power plant, is favorable for optimizing operation of the wind power plant, and can complete prediction of individual wind turbines. Meanwhile, the method has strong expandability, does not depend on a specific wind turbine generator set, and has strong generalization capability.
In addition, the embodiment of the invention also provides a storage medium, wherein the storage medium is stored with a wind power prediction program, and the wind power prediction program realizes the steps of the wind power prediction method when being executed by a processor.
Based on the first embodiment of the wind power prediction method of the present invention, a first embodiment of the wind power prediction device of the present invention is provided, and referring to fig. 6, fig. 6 is a block diagram of the structure of the first embodiment of the wind power prediction device of the present invention.
As shown in fig. 6, the wind power prediction apparatus provided by the embodiment of the present invention includes:
the data processing module 601 is configured to obtain data to be predicted of a wind farm, and perform data preprocessing on the data to be predicted to obtain standard data to be predicted;
the vortex analysis module 602 is used for determining the vortex characteristics of the wind farm according to the standard data to be predicted;
and the power prediction module 603 is configured to predict the standard data to be predicted and the vortex characteristic through a hybrid learning model, so as to obtain wind power of the wind power plant.
According to the embodiment, standard data to be predicted is obtained by acquiring the data to be predicted of the wind power plant and carrying out data preprocessing on the data to be predicted; determining vortex characteristics of the wind power plant according to standard data to be predicted; and predicting the standard data to be predicted and the vortex characteristics through a hybrid learning model to obtain the wind power of the wind power plant. The vortex characteristics of the wind power plant are determined according to the standard data to be predicted of the wind power plant, and the vortex characteristics and the standard data to be predicted are predicted through the hybrid learning model, so that the data dimension of wind power prediction is increased, and the accuracy of wind power prediction is improved.
Further, the vortex analysis module 602 is further configured to process the standard data to be predicted by using a particle swarm algorithm, so as to determine a vortex center position; acquiring a wind field change condition within a preset range of the vortex center position, and optimizing the vortex center position according to the wind field change condition to obtain a final vortex center position; and determining vortex characteristics of each final vortex center position in the wind power plant according to the final vortex center position and the standard data to be predicted.
Further, the vortex analysis module 602 is further configured to predict, by using a hybrid learning model, the standard data to be predicted and the vortex characteristics at the final vortex center position, so as to obtain wind power of each wind turbine in the wind power plant; and determining the wind power of the wind power plant according to the wind power of each wind turbine generator.
Further, the wind power prediction device further includes: a model building module; the model construction module is used for acquiring historical measurement data of the wind power plant, preprocessing the historical measurement data and acquiring standard historical data; performing vortex characteristic analysis on the standard historical data to determine vortex characteristics corresponding to the standard historical data set; generating a training data set and a test data set according to the standard historical data and the vortex characteristics; inputting the training data set into an initial hybrid learning model for training to obtain a premixed learning model; performing precision evaluation on the pre-mixing learning model through a test data set to obtain model precision of the pre-mixing learning model; and generating a hybrid learning model when the model precision of the pre-mixed learning model is higher than the preset precision.
Further, the model construction module is also used for constructing an initial hybrid learning model based on the BP neural network and the support vector machine; and updating and training the BP neural network and the support vector machine through the training data set to obtain a pre-mixing learning model.
Further, the model building module is further used for obtaining a start-stop state of the wind turbine in the wind power plant and generating start-stop parameters of the wind turbine according to the start-stop state of the wind turbine; and generating a training data set and a testing data set according to the standard historical data, the wind turbine generator start-stop parameters and the vortex characteristics.
Further, the model construction module is further configured to return to acquiring historical measurement data of the wind farm when the model precision of the pre-mixing learning model is not higher than a preset precision, and perform preprocessing on the historical measurement data to obtain standard historical data until the model precision of the pre-mixing learning model is higher than the preset precision.
Other embodiments or specific implementation manners of the wind power prediction device of the present invention may refer to the above method embodiments, and are not described herein.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. read-only memory/random-access memory, magnetic disk, optical disk), comprising instructions for causing a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method according to the embodiments of the present invention.
The foregoing description is only of the preferred embodiments of the present invention, and is not intended to limit the scope of the invention, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.
Claims (8)
1. A method for predicting wind power, the method comprising:
obtaining data to be predicted of a wind power plant, and carrying out data preprocessing on the data to be predicted to obtain standard data to be predicted;
determining vortex characteristics of the wind farm according to the standard data to be predicted;
predicting the standard data to be predicted and the vortex characteristics through a hybrid learning model to obtain wind power of the wind power plant;
the step of determining the vortex characteristics of the wind farm according to the standard data to be predicted comprises the following steps:
processing the standard data to be predicted through a particle swarm algorithm to determine the vortex center position;
acquiring a wind field change condition within a preset range of the vortex center position, and optimizing the vortex center position according to the wind field change condition to obtain a final vortex center position;
determining vortex characteristics of each final vortex center position in the wind power plant according to the final vortex center position and the standard data to be predicted;
the step of obtaining the data to be predicted of the wind power plant, and carrying out data preprocessing on the data to be predicted, and before the step of obtaining the standard data to be predicted, further comprises the steps of:
Acquiring historical measurement data of a wind power plant, and preprocessing the historical measurement data to obtain standard historical data;
performing vortex characteristic analysis on the standard historical data to determine vortex characteristics corresponding to the standard historical data set;
generating a training data set and a test data set according to the standard historical data and the vortex characteristics;
inputting the training data set into an initial hybrid learning model for training to obtain a premixed learning model;
performing precision evaluation on the pre-mixing learning model through a test data set to obtain model precision of the pre-mixing learning model;
and generating a hybrid learning model when the model precision of the pre-mixed learning model is higher than the preset precision.
2. The wind power prediction method according to claim 1, wherein the step of predicting the standard data to be predicted and the vortex characteristics by a hybrid learning model to obtain the wind power of the wind farm comprises:
predicting the standard data to be predicted and the vortex characteristics at the final vortex center position through a hybrid learning model to obtain wind power of each wind turbine in the wind power plant;
And determining the wind power of the wind power plant according to the wind power of each wind turbine generator.
3. The wind power prediction method according to claim 1, wherein the step of inputting the training data set into an initial hybrid learning model for training to obtain a premixed learning model includes:
constructing an initial hybrid learning model based on a BP neural network and a support vector machine;
and updating and training the BP neural network and the support vector machine through the training data set to obtain a pre-mixing learning model.
4. The wind power prediction method of claim 1, further comprising, prior to the step of generating a training data set and a test data set from the standard historical data and the vortex characteristics:
acquiring a start-stop state of a wind turbine in a wind power plant, and generating start-stop parameters of the wind turbine according to the start-stop state of the wind turbine;
accordingly, the step of generating a training data set and a test data set from the standard history data and the vortex characteristics comprises:
and generating a training data set and a testing data set according to the standard historical data, the wind turbine generator start-stop parameters and the vortex characteristics.
5. The wind power prediction method according to claim 1, wherein after the step of obtaining the model accuracy of the pre-mix learning model by evaluating the accuracy of the pre-mix learning model by a test data set, further comprising:
and when the model precision of the pre-mixing learning model is not higher than the preset precision, returning to the step of acquiring the historical measurement data of the wind power plant, preprocessing the historical measurement data and obtaining the standard historical data until the model precision of the pre-mixing learning model is higher than the preset precision.
6. A wind power prediction apparatus, characterized in that the wind power prediction apparatus comprises:
the model construction module is used for acquiring historical measurement data of the wind power plant, preprocessing the historical measurement data and acquiring standard historical data; performing vortex characteristic analysis on the standard historical data to determine vortex characteristics corresponding to the standard historical data set; generating a training data set and a test data set according to the standard historical data and the vortex characteristics; inputting the training data set into an initial hybrid learning model for training to obtain a premixed learning model; performing precision evaluation on the pre-mixing learning model through a test data set to obtain model precision of the pre-mixing learning model; generating a hybrid learning model when the model precision of the pre-mixed learning model is higher than a preset precision;
The data processing module is used for acquiring data to be predicted of the wind power plant, and carrying out data preprocessing on the data to be predicted to obtain standard data to be predicted;
the vortex analysis module is used for determining the vortex characteristics of the wind power plant according to the standard data to be predicted;
the power prediction module is used for predicting the standard data to be predicted and the vortex characteristics through the hybrid learning model to obtain wind power of the wind power plant;
the vortex analysis module is further used for processing the standard data to be predicted through a particle swarm algorithm and determining the vortex center position; acquiring a wind field change condition within a preset range of the vortex center position, and optimizing the vortex center position according to the wind field change condition to obtain a final vortex center position; and determining vortex characteristics of each final vortex center position in the wind power plant according to the final vortex center position and the standard data to be predicted.
7. A wind power prediction apparatus, the apparatus comprising: a memory, a processor and a wind power prediction program stored on the memory and executable on the processor, the wind power prediction program configured to implement the steps of the wind power prediction method of any one of claims 1 to 5.
8. A storage medium having stored thereon a wind power prediction program which when executed by a processor implements the steps of the wind power prediction method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311445250.1A CN117175585B (en) | 2023-11-02 | 2023-11-02 | Wind power prediction method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311445250.1A CN117175585B (en) | 2023-11-02 | 2023-11-02 | Wind power prediction method, device, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117175585A CN117175585A (en) | 2023-12-05 |
| CN117175585B true CN117175585B (en) | 2024-03-08 |
Family
ID=88941593
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311445250.1A Active CN117175585B (en) | 2023-11-02 | 2023-11-02 | Wind power prediction method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117175585B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110751326A (en) * | 2019-10-17 | 2020-02-04 | 江苏远致能源科技有限公司 | Photovoltaic day-ahead power prediction method and device and storage medium |
| CN114299377A (en) * | 2021-12-31 | 2022-04-08 | 中山大学 | Vortex identification method and device based on width learning |
| CN114841077A (en) * | 2022-05-24 | 2022-08-02 | 华电电力科学研究院有限公司 | Wind power prediction method, device and medium |
| WO2023124287A1 (en) * | 2021-12-31 | 2023-07-06 | 京东方科技集团股份有限公司 | Wind power output prediction method, electronic device, storage medium, and system |
| CN116415714A (en) * | 2023-02-27 | 2023-07-11 | 国电联合动力技术有限公司 | Wind power prediction method and device, electronic equipment and readable storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104021424B (en) * | 2013-02-28 | 2018-12-07 | 乌托巴斯洞察公司 | Method and apparatus for predicting the output power of the blower in wind field |
-
2023
- 2023-11-02 CN CN202311445250.1A patent/CN117175585B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110751326A (en) * | 2019-10-17 | 2020-02-04 | 江苏远致能源科技有限公司 | Photovoltaic day-ahead power prediction method and device and storage medium |
| CN114299377A (en) * | 2021-12-31 | 2022-04-08 | 中山大学 | Vortex identification method and device based on width learning |
| WO2023124287A1 (en) * | 2021-12-31 | 2023-07-06 | 京东方科技集团股份有限公司 | Wind power output prediction method, electronic device, storage medium, and system |
| CN114841077A (en) * | 2022-05-24 | 2022-08-02 | 华电电力科学研究院有限公司 | Wind power prediction method, device and medium |
| CN116415714A (en) * | 2023-02-27 | 2023-07-11 | 国电联合动力技术有限公司 | Wind power prediction method and device, electronic equipment and readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117175585A (en) | 2023-12-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110761947B (en) | Yaw calibration method and system for wind turbine generator | |
| Kusiak et al. | Short-horizon prediction of wind power: A data-driven approach | |
| CN111441917B (en) | Load estimation method and device for preset component of sector-based wind turbine | |
| CN110210044A (en) | Load prediction method and device of wind generating set | |
| Yin et al. | Big data driven multi-objective predictions for offshore wind farm based on machine learning algorithms | |
| CN112800683B (en) | System short-circuit current level evaluation method and system based on convolutional neural network | |
| Barocio et al. | Online coherency identification and stability condition for large interconnected power systems using an unsupervised data mining technique | |
| CN114341488A (en) | Reinforcement learning based wind turbine yaw offset control | |
| Chen et al. | Comparison of three methods for short-term wind power forecasting | |
| CN113094997B (en) | Wind turbine generator operation simulation method, device, equipment and storage medium | |
| Guo et al. | Wind power assessment based on a WRF wind simulation with developed power curve modeling methods | |
| CN117175585B (en) | Wind power prediction method, device, equipment and storage medium | |
| CN112052604A (en) | Method, system, equipment and readable medium for predicting equivalent fatigue load of fan | |
| CN116885711A (en) | Wind power prediction method, device, equipment and readable storage medium | |
| Li et al. | Nonparametric importance sampling for wind turbine reliability analysis with stochastic computer models | |
| CN112131753A (en) | Method, system and device for evaluating fatigue life of fan and readable medium | |
| CN116432548A (en) | Fusion wind speed calculation method, device, equipment and medium based on transfer learning | |
| CN116245032A (en) | Wind power plant simulation generating capacity correction method and system considering space-time correlation | |
| Manero et al. | Go with the flow: Recurrent networks for wind time series multi-step forecasting | |
| CN116933570B (en) | Method and device for evaluating power generation capacity in wind power plant redevelopment process | |
| Oliveira Schmidt et al. | LSTM Neural Networks Using the SMOTE Algorithm for Wind Turbine Fault Prediction | |
| CN116151130B (en) | Wind power plant maximum frequency damping coefficient calculation method, device, equipment and medium | |
| CN113673167B (en) | Modeling method and system for output power of wind turbine generator based on Gaussian process regression | |
| CN118377067A (en) | Weather early warning method, weather early warning device, computer equipment and storage medium | |
| Cadoret | Operational Modal Analysis (OMA) for wind turbines health monitoring |
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 |