A method and system for predicting the power generation of a wind farm

Abstract

The present invention relates to the technical field of machine learning, and discloses a method and system for predicting the power generation of a wind farm. The method includes obtaining historical data and recent data of the wind farm, and establishing an initial power generation prediction model; selecting historical meteorological factors and historical power generation related to power generation from the historical data; training the initial power generation prediction model using the historical meteorological factors and historical power generation to obtain a power generation prediction model; determining whether the recent data meets the scoring requirements, and making corresponding adjustments to the power generation prediction model; using the power generation prediction model to predict the power generation to obtain power generation data. The present method has the following effects: collecting historical data and recent data of the wind farm, and putting these data into the initial power generation prediction model for training to obtain a complete prediction model. The prediction model can adjust itself in real time according to the recent data, so as to be able to predict the power generation more accurately and timely.

Description

Technical Field

[0001] The present invention relates to the technical field of machine learning, and particularly to a method and system for predicting wind farm power generation. Background Art

[0002] After wind power in China is connected to the power system, due to the characteristics of randomness, intermittency and non-schedulability in the power generation process of wind power, it has a great impact on the safe and economic dispatching of the power system. However, accurate wind power time series prediction will provide important reference basis for the daily dispatching, state estimation and real-time dispatching of the power system, which is beneficial to the optimal allocation of active reserve capacity on the power generation side of the power system and the formulation of demand-side dispatching schemes, and improves the response ability of the power system dispatching to the access of wind power.

[0003] At present, the power generation prediction of most wind farms is based on statistical analysis methods, that is, according to meteorological and wind farm historical data, a power generation prediction model is calculated, such as using multiple linear regression, piecewise linear regression, time series model or grey theory prediction method. However, this type of method ignores the wake effect between units, the differences in wind turbine diameter and hub height, the failure rate and the control factors of wind turbines, and the prediction results are not accurate enough.

[0004] To improve the accuracy of prediction results, there are also methods using power prediction-based methods. Power prediction-based methods usually consider the influence of wind speed on power generation and the wake effect between wind turbines. At present, mainly dynamic time warping algorithm (DTW), Kalman filter algorithm, etc. are used to construct a power generation prediction model, and the future power output is predicted by real-time wind speed, wind direction and atmospheric pressure; or based on the short-term wind speed change, according to the probability distribution of different wind speed segments and the regression curve of wind speed and power, the future power output is calculated. However, the power prediction-based method does not consider the wind energy utilization characteristics of wind turbines. At the same time, if the dynamic time warping algorithm (DTW) is used to construct a power generation prediction model, since there is no objective evaluation standard for the similarity and smoothness of time series, and the power curve of the wind farm is easily affected by the wind power control strategy, the algorithm is prone to outlier; predicting the power output based on the short-term wind speed change, the prediction of short-term wind speed fluctuations is inaccurate and the error is large. Summary of the Invention

[0005] The present invention provides a method and system for predicting wind farm power generation to solve the problem that the prediction model cannot be adjusted in real time according to the objective environment during the prediction process of wind farm power generation, resulting in inaccurate prediction.

[0006] In a first aspect, to solve the above technical problem, the present invention provides a method for predicting wind farm power generation, including:

[0007] Obtain the historical operation data and historical meteorological data of the wind farm, as well as the recent data of the wind farm, and establish an initial power generation prediction model; the recent data includes recent meteorological data, recent operation data, and recent output log data within a preset time period from the current moment;

[0008] Select historical meteorological factors related to the power generation of the wind farm from the historical meteorological data;

[0009] Select the historical power generation related to the historical meteorological factors from the historical operation data;

[0010] Use the historical meteorological factors and the historical power generation to train the initial power generation prediction model to obtain a first power generation prediction model;

[0011] Perform multi-scale feature extraction on the recent data to obtain multi-scale feature data, and input the multi-scale feature data into a multi-scale normalizing flow model to obtain multi-scale normalizing flow data;

[0012] Input the multi-scale normalizing flow data into a hypersphere support vector machine model for multi-dimensional time series data classification to obtain classified output time series data;

[0013] According to the output time series data, calculate the contribution degree score to obtain the real-time contribution degree score of each output time series data;

[0014] When it is determined that the real-time contribution degree score is greater than a preset score threshold, then use the recent data to train the first power generation prediction model to obtain a second power generation prediction model; when it is determined that the real-time contribution degree score is less than or equal to the preset score threshold, then directly use the output of the first power generation prediction model as the second power generation prediction model;

[0015] Use the second power generation prediction model to predict the future power generation of the wind farm to obtain wind farm power generation data.

[0016] In an alternative embodiment, the inputting the multi-scale normalizing flow data into a hypersphere support vector machine model for multi-dimensional time series data classification to obtain classified output time series data includes:

[0017] Construct a hypersphere in the feature space according to the multi-scale normalizing flow data to obtain hypersphere data;

[0018] Perform a data projection operation according to the multi-scale normalizing flow data and the hypersphere data and calculate the distance between the data point and the center of the sphere to obtain the center-of-sphere distance data;

[0019] Based on the center distance data of the sphere, multi-dimensional time-series data is classified to obtain the output time-series data after classification.

[0020] In an alternative embodiment, the calculating the real-time contribution score for each output time-series data according to the output time-series data includes:

[0021] Calculating the mean value of the core time-series data according to the output time-series data to obtain the core center point;

[0022] Calculating the Euclidean distance between each time-series data and the core center point according to the output time-series data and the core center point to obtain the proximity distance;

[0023] Calculating the contribution score of the time-series data according to the proximity distance to obtain the real-time contribution score for each time-series data.

[0024] In an alternative embodiment, the historical operation data includes the output, power, and corresponding power generation of the wind farm units; the historical meteorological data includes the historical wind speed and the corresponding wind direction;

[0025] The initial power generation prediction model includes a power prediction model based on physical factors and a statistical model based on statistical analysis methods; wherein, the power prediction model includes a power function of the power generation power with respect to time and wind speed; the statistical model includes a statistical function of the output of the wind farm units with respect to wind speed.

[0026] In an alternative embodiment, the selecting the historical meteorological factors related to the wind farm power generation from the historical meteorological data includes:

[0027] Setting a correlation coefficient threshold for the historical meteorological data and the wind farm power generation;

[0028] Calculating the correlation coefficient between the historical meteorological data and the wind farm power generation;

[0029] Selecting the historical meteorological data corresponding to the correlation coefficient greater than the correlation coefficient threshold and outputting it as the historical meteorological factor.

[0030] In an alternative embodiment, the training the initial power generation prediction model using the historical meteorological factors and the historical power generation to obtain the first power generation prediction model includes:

[0031] Using the polynomial fitting method to train the power prediction model according to the historical power generation to obtain the target power prediction model;

[0032] Using the least squares method to train the statistical model according to the historical meteorological factors to obtain the target statistical model;

[0033] Perform a model fusion operation on the target power prediction model and the target statistical model to obtain the first power generation prediction model.

[0034] In an alternative embodiment, calculating the correlation coefficient between the historical meteorological data and the wind farm power generation includes:

[0035] The correlation coefficient is calculated according to the following formula

[0036]

[0037] where represents the correlation coefficient, represents the wind speed at time, represents the average value of the wind speed in the time period of the historical meteorological data, represents at the wind farm power generation at time, represents the average value of the wind farm power generation in the time period of the historical meteorological data, represents the number of all time periods recorded in the historical meteorological factors.

[0038] In an alternative embodiment, using the polynomial fitting method to train the power prediction model according to the historical power generation to obtain the target power prediction model includes:

[0039] Calculate the power function according to the historical power generation;

[0040] Substitute the power function into the power prediction model to obtain the target power prediction model;

[0041] The power function is calculated according to the following formula

[0042]

[0043] where represents the power function at time, represents at the historical power generation at time, represents at the wind speed at time.

[0044] In an alternative embodiment, using the least squares method to train the statistical model according to the historical meteorological factors to obtain the target statistical model includes:

[0045] Calculate the statistical function according to the historical meteorological factors;

[0046] Substitute the statistical function into the statistical model to obtain the target statistical model;

[0047] The statistical function is calculated according to the following formula,

[0048]

[0049] where, represents the statistical function at time, represents at the power of the wind speed at time, represents the average value of the wind speed in the time period of the historical meteorological factors, represents the number of all time periods recorded in the historical meteorological factors.

[0050] In a second aspect, the present invention provides a wind farm power generation prediction system, which is characterized by including:

[0051] A data acquisition module, configured to acquire historical operation data and historical meteorological data of a wind farm, as well as recent data of the wind farm, and establish an initial power generation prediction model;

[0052] A historical meteorological factor selection module, configured to select historical meteorological factors related to the power generation of the wind farm from the historical meteorological data;

[0053] A historical power generation selection module, configured to select historical power generation related to the historical meteorological factors from the historical operation data;

[0054] A model training module, configured to train the initial power generation prediction model using the historical meteorological factors and the historical power generation to obtain a first power generation prediction model;

[0055] A model optimization module, configured to perform multi-scale feature extraction on the recent data to obtain multi-scale feature data, and input the multi-scale feature data into a multi-scale normalizing flow model to obtain multi-scale normalizing flow data; input the multi-scale normalizing flow data into a hypersphere support vector machine model for multi-dimensional time series data classification to obtain classified output time series data; calculate a contribution score according to the output time series data to obtain the real-time contribution score of each output time series data; when it is determined that the real-time contribution score is greater than a preset score threshold, then train the first power generation prediction model using the recent data to obtain a second power generation prediction model; when it is determined that the real-time contribution score is less than or equal to the preset score threshold, then directly output the first power generation prediction model as the second power generation prediction model;

[0056] A model prediction module, configured to use the second power generation prediction model to predict the future power generation of a wind farm, and obtain wind farm power generation data.

[0057] In a third aspect, the present invention further provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the wind farm power generation prediction method described in any one of the above is implemented.

[0058] In a fourth aspect, the present invention further provides a computer-readable storage medium, which includes a stored computer program. When the computer program runs, it controls the device where the computer-readable storage medium is located to execute the wind farm power generation prediction method described in any one of the above.

[0059] Compared with the prior art, the present invention has the following beneficial effects:

[0060] The present invention provides a wind farm power generation prediction method, including obtaining historical operation data and historical meteorological data of the wind farm, as well as recent data of the wind farm, and establishing an initial power generation prediction model; selecting historical meteorological factors related to the power generation of the wind farm from the historical meteorological data; selecting historical power generation related to the historical meteorological factors from the historical operation data; using the historical meteorological factors and the historical power generation to train the initial power generation prediction model to obtain a first power generation prediction model; performing multi-scale feature extraction on the recent data to obtain multi-scale feature data, and inputting the multi-scale feature data into a multi-scale normalizing flow model to obtain multi-scale normalizing flow data; inputting the multi-scale normalizing flow data into a hypersphere support vector machine model for multi-dimensional time series data classification to obtain classified output time series data; calculating contribution degree scores according to the output time series data to obtain real-time contribution degree scores of each output time series data; when it is determined that the real-time contribution degree score is greater than a preset score threshold, training the first power generation prediction model with the recent data to obtain a second power generation prediction model; when it is determined that the real-time contribution degree score is less than or equal to the preset score threshold, directly using the output of the first power generation prediction model as the second power generation prediction model; using the second power generation prediction model to predict the future power generation of the wind farm to obtain wind farm power generation data.

[0061] The method obtains an initial power generation prediction model by integrating a power model and a statistical model. Meanwhile, various historical data and recent data of the wind farm are collected and put into the initial power generation prediction model for training to obtain a complete prediction model. Among them, the hypersphere support vector machine has good performance in processing high-dimensional data and can improve the accuracy of classification. According to the classified output time-series data, the contribution score is calculated to obtain the real-time contribution score of each output time-series data, which can identify the factors that have the greatest impact on power generation. If the real-time contribution score is greater than the preset score threshold, it indicates that the recent data has a significant impact on the prediction ability of the model, and these data need to be used to update the first power generation prediction model to obtain the second power generation prediction model, so that it can be adjusted in real time according to the recent data, and thus can more accurately and timely predict the future power generation of the wind farm. If the real-time contribution score is less than or equal to the preset score threshold, it indicates that the recent data has no significant impact on the prediction ability of the model, and there is no need to update the prediction model, which does not occupy additional memory and operation memory of the system and is conducive to the long-term stable and reliable operation of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 is a schematic flowchart of the wind farm power generation prediction method provided by the first embodiment of the present invention;

[0063] Figure 2 is a schematic structural diagram of the wind farm power generation prediction system provided by the second embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0064] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0065] Refer to Figure 1 , the first embodiment of the present invention provides a wind farm power generation prediction method, including the following steps:

[0066] S1. Obtain the historical operation data and historical meteorological data of the wind farm, as well as the recent data of the wind farm, and establish an initial power generation prediction model;

[0067] S2. Select historical meteorological factors related to the power generation of the wind farm from the historical meteorological data;

[0068] S3. Select historical power generation related to the historical meteorological factors from the historical operation data;

[0069] S4. Use the historical meteorological factors and the historical power generation to train the initial power generation prediction model to obtain a first power generation prediction model;

[0070] S5. Perform multi-scale feature extraction on the recent data to obtain multi-scale feature data, and input the multi-scale feature data into a multi-scale normalizing flow model to obtain multi-scale normalizing flow data;

[0071] S6. Input the multi-scale normalizing flow data into a hypersphere support vector machine model for multi-dimensional time series data classification to obtain classified output time series data;

[0072] S7. Calculate contribution score based on the output time series data to obtain the real-time contribution score of each output time series data;

[0073] S8. When it is determined that the real-time contribution score is greater than a preset score threshold, use the recent data to train the first power generation prediction model to obtain a second power generation prediction model; when it is determined that the real-time contribution score is less than or equal to the preset score threshold, directly use the output of the first power generation prediction model as the second power generation prediction model;

[0074] S9. Use the second power generation prediction model to predict the future power generation of the wind farm to obtain wind farm power generation data.

[0075] It should be noted that currently, the power generation prediction methods for wind farms mainly include those based on statistical analysis methods and power prediction methods. For the statistical analysis method, that is, according to meteorological and wind farm historical data, calculate the power generation prediction model, such as using multiple linear regression, piecewise linear regression, time series model or grey theory prediction method. However, this type of method ignores the wake effect between units, the differences in wind turbine diameters and hub heights, the failure rate and wind turbine control factors, and the prediction results are not accurate enough. The power prediction-based methods usually consider the influence of wind speed on power generation and the wake effect between wind turbines. Currently, mainly use dynamic time warping algorithm, Kalman filter algorithm, etc. to construct the power generation prediction model, and predict the future power output from real-time wind speed, wind direction and atmospheric pressure; or based on the short-term wind speed change, according to the probability distribution of different wind speed segments and the regression curve of wind speed and power, calculate the future power output. However, this method does not consider the wind energy utilization characteristics of wind turbines. At the same time, if using the dynamic time warping algorithm to construct the power generation prediction model, due to the lack of an objective evaluation standard for the similarity and smoothness of time series, and the power curve of the wind farm is easily affected by the wind power control strategy, the algorithm is prone to error dispersion; predicting the power output based on short-term wind speed changes, the prediction of short-term wind speed fluctuations is inaccurate and the error is large.

[0076] In the present invention, a power model and a statistical model are fused to obtain an initial power generation prediction model. Meanwhile, various historical data and recent data of the wind farm are collected and put into the initial power generation prediction model for training to obtain a complete prediction model. This prediction model can adjust itself in real time according to recent data, so as to more accurately and timely predict the future power generation of the wind farm.

[0077] In step S1, first, the historical operation data and historical meteorological data of the current wind farm, as well as the recent data of the wind farm, are obtained. Then, a suitable machine learning algorithm is selected to establish an initial power generation prediction model.

[0078] Among them, the historical operation data and historical meteorological data of the wind farm, as well as the recent data of the wind farm, include that the historical operation data includes the output, power of the wind farm units, and the corresponding power generation; the historical meteorological data includes the historical wind speed and the corresponding wind direction; the recent data includes the recent meteorological data, recent operation data, and recent output log data within a preset time period from the current moment. Exemplarily, the historical operation data refers to the operation records of the wind farm in the past period of time, including the output, power of the wind farm units, and the corresponding power generation, such as data in the past few years or months, specifically depending on the time span required by the model. The historical meteorological data refers to the meteorological condition records of the wind farm in the past period of time, including the wind speed and wind direction. Similar to the historical operation data, the historical meteorological data refers to data in the past few years or months.

[0079] It should be noted that the recent data refers to the data within a preset time period from the current moment, which is used to capture the latest operation and meteorological conditions to improve the accuracy of the prediction model. This preset time period can be several days, weeks or a month, specifically depending on the prediction requirements of the model and the availability of data. Among them, the recent meteorological data includes the recent wind speed, wind direction, temperature, humidity and other meteorological parameters, the recent operation data includes the output and power of the wind farm units in the recent period of time, and the recent output log data records the output conditions of the units in the recent period of time, including output time, faults, maintenance and other events.

[0080] Specifically, first, the historical operation data and historical meteorological data of the current wind farm, as well as the recent data of the current wind farm, need to be obtained through the records in the file materials or database. Among them, the recent data of the current wind farm can also be obtained by real-time detection using various instruments installed in the wind farm. Then, according to the specific data types and characteristics of the obtained data, a suitable machine learning algorithm is selected to establish an initial power generation prediction model.

[0081] It should be noted that the specific data type of the data needs to be considered when selecting an appropriate machine learning algorithm. For example, weather data includes satellite cloud map data, so an algorithm commonly used for image data can be selected. Commonly used machine learning algorithms mainly focus on the field of computer vision, especially deep learning algorithms. Some common choices: Convolutional Neural Network, a neural network that can automatically learn features from images, widely used in image classification, object detection, and image segmentation tasks; Scale-Invariant Feature Transform, a feature detection algorithm used to detect and describe local features in images, which is invariant to scaling, rotation, and illumination changes, and is suitable for tasks such as image matching and object recognition; Histogram of Oriented Gradients Figure 1 a feature extraction technique used to detect objects in images by calculating the distribution of gradient directions in the image, and is commonly used in scenarios such as dynamic detection.

[0082] In one embodiment, the initial power generation prediction model includes a power prediction model based on physical factors and a statistical model based on statistical analysis methods; wherein, the power prediction model includes a power function of the power generation power with respect to time and wind speed; the statistical model includes a statistical function of the output of the wind farm units with respect to wind speed.

[0083] In step S2, historical meteorological factors related to the current power generation of the wind farm need to be selected from the historical meteorological data as the data for subsequent training of the model.

[0084] In one embodiment, the selection of historical meteorological factors related to the power generation of the wind farm from the historical meteorological data includes: setting a correlation coefficient threshold for the historical meteorological data and the power generation of the wind farm; calculating the correlation coefficient between the historical meteorological data and the power generation of the wind farm; selecting the historical meteorological data corresponding to the correlation coefficient greater than the correlation coefficient threshold, and outputting it as the historical meteorological factor.

[0085] Specifically, first, according to the relationship between the historical meteorological data and the power generation of the wind farm, a suitable statistical index of the degree of correlation is selected, and a corresponding correlation coefficient threshold is set, that is, if it exceeds a certain interval, it can be considered that the historical meteorological data and the power generation of the wind farm are relevant. Then, the correlation coefficient is calculated according to the formula. Finally, from all the historical meteorological data, the historical meteorological data corresponding to the correlation coefficient greater than the correlation coefficient threshold is selected and output as the historical meteorological factor to be used as the data for subsequent training of the model.

[0086] In one embodiment, the calculation of the correlation coefficient between the historical meteorological data and the power generation of the wind farm includes: calculating the correlation coefficient according to the following formula,

[0087]

[0088] Among them, represents the said correlation coefficient, represents at the wind speed at the moment, represents the average value of the wind speed during the time period of the said historical meteorological data, represents at the power generation of the wind farm at the moment, represents the average value of the power generation of the wind farm during the time period of the said historical meteorological data, represents the total number of time periods recorded in the said historical meteorological factors. The said correlation coefficient is the Pearson correlation coefficient, which is mainly used to measure the degree of linear correlation between two variables, and its value ranges from -1 to 1. When the value of the correlation coefficient is positive, it indicates that the two variables are monotonically positively correlated; when the value of the correlation coefficient is negative, it indicates that the two variables are monotonically negatively correlated; the closer the value of the correlation coefficient is to 0, the less monotonous relationship exists between the two variables.

[0089] In another embodiment, the Spearman correlation coefficient is selected to judge the correlation between the said historical meteorological data and the power generation of the wind farm. The said Spearman correlation coefficient is calculated according to the following formula

[0090]

[0091] Among them, represents the said Spearman correlation coefficient, represents the difference in the ranks of each pair of observed values, that is, the difference in the positions of the corresponding observed values of the two variables in the sorted list, represents the number of observed values. The value range of the said Spearman correlation coefficient is between -1 and 1, and the larger its absolute value, the stronger the rank correlation between the two variables. It should be noted that the said Spearman correlation coefficient is calculated based on ranks, so its value is not affected by the distribution of the original data and has better robustness to outliers. This makes it more applicable than the Pearson correlation coefficient when dealing with datasets that do not meet the normal distribution assumption or contain outliers.

[0092] In step S3, it is necessary to select the historical power generation related to the said historical meteorological factors from the said historical operation data as the data for subsequent training of the model.

[0093] Specifically, the historical operation data includes historical power generation, and the wind farm will generate different amounts of power due to the historical meteorological factors. Therefore, selecting the historical power generation related to the historical meteorological factors from the historical operation data is to select the representative wind farm power generation under different historical meteorological factors, such as different wind speeds and different weather conditions. For multiple power generations under approximate historical meteorological factors, any one of the power generations can be selected as the power generation under this historical meteorological factor. Finally, the wind farm power generations under all different historical meteorological factors are combined to form the historical power generation.

[0094] It should be noted that the historical meteorological factors have multiple aspects. To determine whether two different historical meteorological factors are approximate, it is necessary to comprehensively consider indicators such as wind speed, time, location, humidity, and temperature, and set a certain interval for each indicator. Indicators within the same interval can be considered approximate. Further, only two different historical meteorological factors with all indicators within the same interval can be considered approximate.

[0095] In step S4, the initial power generation prediction model is trained using the historical meteorological factors and the historical power generation to obtain the first power generation prediction model.

[0096] In one implementation manner, the step of using the historical meteorological factors and the historical power generation to train the initial power generation prediction model to obtain the first power generation prediction model includes the following steps:

[0097] S41, using the polynomial fitting method, training the power prediction model according to the historical power generation to obtain the target power prediction model.

[0098] S42, using the least squares method, training the statistical model according to the historical meteorological factors to obtain the target statistical model.

[0099] S43, performing a model fusion operation on the target power prediction model and the target statistical model to obtain the first power generation prediction model.

[0100] In step S41, using the polynomial fitting method, substituting the historical power generation into the power prediction model for training to obtain the target power prediction model.

[0101] It should be noted that the core of the power prediction model is the power function. By determining and calculating the power function, a specific calculation mode can be set for the power prediction model, thereby obtaining the target power prediction model, enabling the model to perform accurate analysis and prediction based on the input data.

[0102] In one implementation, the power prediction model is trained according to the historical power generation using the polynomial fitting method to obtain a target power prediction model, including: calculating the power function according to the historical power generation; substituting the power function into the power prediction model to obtain the target power prediction model; calculating the power function according to the following formula,

[0103]

[0104] where, represents the power function at the time of ; represents the historical power generation at the time of ; represents the wind speed at the time of . Using the polynomial fitting method to calculate the power function can well simulate the power generation of the wind farm in the actual process. At the same time, as the power of the polynomial fitting increases, theoretically the fitting accuracy can be improved, but the calculation amount will also increase significantly; therefore, when fitting, only fit to the cubic polynomial, which avoids a large increase in the calculation amount and also maintains an acceptable accuracy.

[0105] In step S42, using the least squares method, the historical meteorological factors are substituted into the statistical model for training to obtain a target statistical model.

[0106] It should be noted that the core of the statistical model is the statistical function. By determining and calculating the statistical function, a specific statistical mode can be set for the statistical model, thereby obtaining the target statistical model, enabling the model to perform accurate statistical analysis on the input data.

[0107] In one implementation, the statistical model is trained according to the historical meteorological factors using the least squares method to obtain a target statistical model, including: calculating the statistical function according to the historical meteorological factors; substituting the statistical function into the statistical model to obtain the target statistical model; calculating the statistical function according to the following formula,

[0108]

[0109] where, represents the statistical function at the time of ; represents the power of the wind speed at the time of ; represents the average value of the wind speed in the time period of the historical meteorological factors, Represents the total number of time periods recorded in the historical meteorological factors. The statistical model obtained using the least squares method can well reflect the overall trend of the data and can also weaken the impact of anomalies in the case of noise or outliers between data points. Such a statistical model can significantly improve the accuracy for data analysis and prediction.

[0110] In step S43, first obtain the previously obtained target power prediction model and the target statistical model, then perform a model fusion operation on the two models, integrate them into a new model, and output it as the first power generation prediction model.

[0111] Specifically, perform a weighted average on the two functions of the two models to obtain a new weighted average function, then select a suitable method to stack the structural parts of the two models to obtain a new model structure, and finally put the new function into the new model structure to obtain the first power generation prediction model. By fusing two types of models, their advantages can be combined, thereby improving the overall performance and accuracy, and being able to better handle outliers and missing values.

[0112] It should be noted that when performing a weighted average, an appropriate weight needs to be selected according to the actual situation to ensure that the real situation can be accurately simulated; similarly, when stacking the structural parts of the two models, a suitable method should also be selected, such as single-layer stacking, cross-validation stacking, to ensure the integrity of the model structure.

[0113] In step S5, perform multi-scale feature extraction on the recent data to obtain multi-scale feature data, and input the multi-scale feature data into the multi-scale normalizing flow model to obtain multi-scale normalizing flow data.

[0114] In one implementation, extract the features in the spatial dimension of the recent data through a convolutional neural network to obtain wind power spatial features; according to the recent data, extract the features in the time dimension through a gated recurrent unit to obtain output time features; according to the wind power spatial features and the output time features, perform multi-scale convolution operations to obtain multi-scale feature data.

[0115] It should be noted that the Convolutional Neural Network (CNN) is a deep learning model. CNN can be used to extract features in the spatial dimension, and these features help capture the spatial distribution characteristics of meteorological conditions such as wind speed and wind direction at different positions in the wind farm. CNN processes the input recent data (meteorological condition data, spatial distribution map of wind farm data) through its convolutional layers to extract the spatial features of the wind farm. These features include local changes in wind speed, consistency of wind direction, etc., which are important factors affecting the power generation of the wind farm. By using convolutional kernels of different sizes and shapes, CNN can detect spatial features at different scales, which is crucial for understanding the complex meteorological patterns inside the wind farm. In multi-scale convolutional operations, CNN can simultaneously use multiple convolutional kernels of different sizes to capture spatial features at different scales, which is beneficial for identifying small-scale wind speed changes locally and large-scale wind speed patterns across the entire wind farm. In the final stage of CNN, the extracted multi-scale spatial features are fused through the Fully Connected Layers to form a comprehensive understanding of the spatial features of the wind farm, and these features can then be input into subsequent models together with other time-dimensional features for further processing.

[0116] Furthermore, based on the multi-scale feature data, a scale unification operation is performed to obtain unified scale data; the unified scale data is subjected to a reversible mapping operation to obtain low-dimensional feature data; based on the unified scale data, a multi-scale data augmentation operation is performed to obtain augmented distribution data; based on the low-dimensional feature data and the augmented distribution data, a multi-distribution fusion operation is performed to obtain multi-scale normalizing flow data.

[0117] Among them, the scale unification operation refers to when processing input data, converting features with different scales or units into the same scale so that subsequent calculation and processing processes can be carried out on a unified basis. Specifically, different data features may have different ranges, units, or distributions in their original forms. Through the scale unification operation, these features can be mapped into the same scale range through standardization, normalization, or other transformation methods, enabling the model to more efficiently utilize these features during training or inference, and avoiding numerical biases or instabilities caused by different scales. In the contribution degree scoring of recent data, the scale unification operation reduces the sensitivity of the model to outliers and noise because the impact of outliers is weakened after standardization, thereby enhancing the robustness of the model and further improving the accuracy and performance of the overall evaluation.

[0118] Specifically, the calculation formula for the reversible mapping operation is:

[0119] ;

[0120] Among them, represents the uniformly scaled data of the input, such as the data obtained by performing a scaling operation on meteorological condition data, wind farm data, and output log data; represents the latent space data after being transformed by the multi-scale normalizing flow model; represents the invertible mapping function of the normalizing flow model; represents the inverse operation of the mapping, that is, mapping the representation in the latent space back to the input space ; represents the layer mapping operation of the flow model, usually an invertible affine transformation; represents the corresponding inverse transformation.

[0121] It should be noted that the invertible mapping operation refers to mapping high-dimensional data to a low-dimensional space through a series of transformations while ensuring the reversibility of the mapping process, that is, the original high-dimensional data can be accurately restored from the low-dimensional data. The core of this operation lies in minimizing information loss during the process of reducing the data dimension and ensuring that the important features of the original data are retained. Exemplarily, in the embodiments of the present invention, the invertible mapping operation adopts the Flow Model technology and is implemented through continuous bidirectional transformations (the forward transformation is used for dimension reduction, and the reverse transformation is used for restoration). In the contribution degree scoring of recent data, using the invertible mapping operation can effectively compress the features of complex multi-dimensional time series data into low-dimensional data to reduce the computational complexity, and at the same time, it can be restored when needed for more accurate contribution degree evaluation and feature analysis. This operation ensures the efficiency and accuracy of data processing.

[0122] In an alternative embodiment, the multi-scale data augmentation operation includes:

[0123] Based on the uniformly scaled data, perform probability density estimation of the data distribution to obtain probability distribution data;

[0124] According to the probability distribution data, perform layer-by-layer normalization operation to obtain augmented distribution data;

[0125] The calculation formula for the probability density estimation is as follows,

[0126] ,

[0127] Among them, represents the uniformly scaled data of the input, such as the data obtained by performing a scaling operation on meteorological condition data, wind farm data, and output log data; represents the latent space data after being transformed by the multi-scale normalizing flow model; Represents the invertible mapping function of the normalizing flow model; Represents The probability density in the latent space, generally assumed To follow a simple known distribution, such as the standard normal distribution; Represents the determinant of the Jacobian matrix, which is used to describe the change in the volume of the feature space during the mapping process.

[0128] It should be noted that the multi-scale data augmentation operation includes two steps: probability density estimation and layer-by-layer normalization. The probability density estimation step is used to estimate the probability distribution of data at different scales. By using statistical methods to model the distribution of data, the occurrence probability of each feature value at different scales is determined. This helps to identify abnormal data points because abnormal points usually have low density values in the probability distribution. Through accurate probability density estimation, the model can effectively distinguish normal and abnormal data patterns and improve the accuracy of evaluation. The layer-by-layer normalization step normalizes each layer of the multi-scale data to ensure that different features are on the same numerical scale. Layer-by-layer normalization can avoid the adverse effects of differences in the order of magnitude of feature values on the model learning process and enhance the stability and robustness of the model when processing data at different scales. This step adjusts the data to a unified range through standardization or normalization formulas to ensure that in subsequent multi-distribution fusion operations, the model can better utilize these enhanced feature data.

[0129] Specifically, the calculation formula for the multi-distribution fusion operation is:

[0130] ;

[0131] Where, Represents the fused multi-scale normalizing flow data; Represents the Probability distribution of the \(i\)-th feature dimension; Represents the weight parameter of each distribution Used to adjust the contribution of each distribution during the fusion process; Is the number of distributions to be fused.

[0132] It should be noted that the multi-distribution fusion operation fuses the data distributions of multiple different feature dimensions in a weighted manner to form a comprehensive probability distribution, representing the final multi-scale normalizing flow data. In this operation, the probability distribution Of each feature dimension Adjusts its contribution to the overall distribution according to its weight parameter , which ensures that the influence of different feature dimensions can be dynamically adjusted during the fusion process. The core function of the multi-distribution fusion operation is to integrate the information of multiple feature dimensions to generate a unified probability model. This model can more accurately represent the overall data distribution, help capture cross-dimensional correlation features during contribution evaluation, and improve the accuracy of contribution evaluation. In addition, the weight parameter This makes the operation flexible and can adaptively adjust the contribution of each distribution according to the importance of different features, further improving the evaluation accuracy.

[0133] In step S6, the multi-scale normalized flow data is input into a hypersphere support vector machine model to perform multi-dimensional time series data classification to obtain classified output time series data, including:

[0134] According to the multi-scale normalized flow data, a hypersphere is constructed in a feature space to obtain hypersphere data;

[0135] According to the multi-scale normalized flow data and the hypersphere data, a data projection operation is performed and the distance between the data point and the center of the sphere is calculated to obtain the center-of-sphere distance data;

[0136] According to the ball center distance data, multi-dimensional time series data classification is performed to obtain classified output time series data.

[0137] Among them, in machine learning, feature space refers to the high-dimensional space to which the original data is mapped through some transformation (such as kernel function). In this space, data points can be more easily distinguished by linear or nonlinear models. Through the kernel function, the original data can be mapped to a high-dimensional feature space, so that the nonlinear relationship of the data in the original space becomes linearly separable in the feature space. In high-dimensional space, a hypersphere is a geometric shape, similar to a circle in two-dimensional space or a ball in three-dimensional space. The center of the hypersphere is a certain center point of all data points in the feature space (such as the mean point), and the radius is determined by the distance between the data point and the center point. In the feature space, a hypersphere is constructed by calculating the center and radius of the data point so that the core data point is contained inside the hypersphere as much as possible. The sphere center distance refers to the distance between the data point and the center point of the hypersphere in the feature space. This distance can be used to measure the closeness of the data point to the core data distribution. Hypersphere support vector machine is a support vector machine model based on hypersphere geometry, which is used for classification tasks, especially in multidimensional time series data. It can effectively process multi-dimensional time series data, identify the relative distribution of data points and core data, and has good robustness and classification performance.

[0138] In step S7, a contribution score calculation is performed based on the output time series data to obtain a real-time contribution score for each output time series data, including:

[0139] Calculate the mean value of the core timing data based on the output timing data to obtain the core center point;

[0140] Calculate the Euclidean distance between each timing data and the core center point according to the output timing data and the core center point to obtain the proximity distance;

[0141] Calculate the contribution score of the timing data according to the proximity distance to obtain the real-time contribution score of each timing data.

[0142] It should be noted that the core center point is obtained by calculating the mean value of the core timing data. The core center point represents the central position of the core data and is used to calculate the proximity of each timing data point to the core data center in the subsequent calculation. The proximity distance is obtained by calculating the Euclidean distance between each timing data point and the core center point. The proximity distance is used to measure the proximity of each timing data point to the core data center. The smaller the proximity distance, the closer the data point is to the core data distribution. The real-time contribution score is obtained by normalizing the calculation according to the proximity distance. The smaller the proximity distance, the higher the real-time contribution score. The real-time contribution score is used to quantify the contribution degree of each timing data point. The higher the score, the closer the data point is to the core point, the higher its contribution degree, and the greater the impact on the final prediction. These steps work together to effectively identify the contributing data points close to the core data distribution and provide a real-time contribution score for each timing data point to help evaluate whether the contribution of recent data is significant.

[0143] In step S8, when it is determined that the real-time contribution score is greater than the preset score threshold, the first power generation prediction model is trained using the recent data to obtain the second power generation prediction model; when it is determined that the real-time contribution score is less than or equal to the preset score threshold, the output of the first power generation prediction model is directly used as the second power generation prediction model.

[0144] Among them, if the real-time contribution score is greater than the preset score threshold, it means that the recent data has a significant impact on the prediction ability of the model, and these data need to be used to update the first power generation prediction model to obtain the second power generation prediction model. If the score is less than or equal to the preset threshold, it means that the recent data has little impact on the prediction ability of the model, and the first power generation prediction model can be directly used as the second model. Exemplarily, when the operation of the wind farm has a general requirement for prediction accuracy, a relatively high threshold, such as 0.8 or 0.9, can be set to ensure that the model is updated only when the recent data has a significant impact on the prediction; when the operation of the wind farm has a very high requirement for prediction real-time performance, a relatively low threshold, such as 0.5 or 0.6, can be set to update the model more frequently to capture the latest data changes.

[0145] In step S9, the trained second power generation prediction model is used to predict the future power generation of the current wind farm, so as to obtain the wind farm power generation data, and then the data is output and saved.

[0146] To facilitate the understanding of the present invention, the working process of the present invention is described below by taking a relatively common scenario as an example.

[0147] In this embodiment, it is planned to implement a system capable of predicting the future power generation of a wind farm. The system aims to accurately and real-time predict the power generation of the wind farm and make corresponding adjustments to itself according to the changes of objective factors such as the wind farm and meteorology.

[0148] Step 1: Obtain the historical operation data, historical meteorological data and recent data of the wind farm from the relevant database, and select a suitable initial power generation prediction model.

[0149] Step 2: Select the historical meteorological factors related to the wind farm power generation from the historical meteorological data and record them.

[0150] Step 3: Select the historical power generation related to the historical meteorological factors from the historical operation data and record them.

[0151] Step 4: Use the historical meteorological factors and historical power generation to train the initial power generation prediction model to obtain the first power generation prediction model.

[0152] Step 5: Judge whether the recent data meets the scoring requirements, and adjust the first power generation prediction model according to the corresponding results to obtain the second power generation prediction model.

[0153] Step 6: Use the second power generation prediction model to predict the future power generation of the wind farm to obtain the wind farm power generation data.

[0154] Through the above steps, the wind farm power generation prediction system can accurately and timely predict the future power generation of the wind farm and make corresponding adjustments according to the real-time data.

[0155] In summary, the present method obtains an initial power generation prediction model by fusing a power model and a statistical model. Meanwhile, various historical data and recent data of the wind farm are collected and put into the initial power generation prediction model for training to obtain a complete prediction model. Among them, the hypersphere support vector machine has good performance in processing high-dimensional data and can improve the accuracy of classification. According to the classified output time series data, the contribution score is calculated to obtain the real-time contribution score of each output time series data, which can identify the factors that have the greatest impact on power generation. If the real-time contribution score is greater than the preset score threshold, it indicates that the recent data has a significant impact on the prediction ability of the model, and these data need to be used to update the first power generation prediction model to obtain the second power generation prediction model, so that the model can be adjusted in real time according to the recent data, and thus can more accurately and timely predict the future power generation of the wind farm.

[0156] Referring to Figure 2 , the second embodiment of the present invention provides a wind farm power generation prediction system, including:

[0157] A data acquisition module, configured to acquire the historical operation data and historical meteorological data of the wind farm, as well as the recent data of the wind farm, and establish an initial power generation prediction model;

[0158] A historical meteorological factor selection module, configured to select historical meteorological factors related to the power generation of the wind farm from the historical meteorological data;

[0159] A historical power generation selection module, configured to select historical power generation related to the historical meteorological factors from the historical operation data;

[0160] A model training module, configured to train the initial power generation prediction model using the historical meteorological factors and the historical power generation to obtain a first power generation prediction model;

[0161] A model optimization module, configured to perform multi-scale feature extraction on the recent data to obtain multi-scale feature data, and input the multi-scale feature data into a multi-scale normalizing flow model to obtain multi-scale normalizing flow data; input the multi-scale normalizing flow data into a hypersphere support vector machine model for multi-dimensional time series data classification to obtain classified output time series data; calculate the contribution score according to the output time series data to obtain the real-time contribution score of each output time series data; when it is determined that the real-time contribution score is greater than a preset score threshold, then use the recent data to train the first power generation prediction model to obtain a second power generation prediction model; when it is determined that the real-time contribution score is less than or equal to the preset score threshold, then directly output the first power generation prediction model as the second power generation prediction model;

[0162] A model prediction module, configured to use the second power generation prediction model to predict the future power generation of a wind farm, so as to obtain wind farm power generation data.

[0163] It should be noted that the wind farm power generation prediction system provided in the embodiments of the present invention is used to execute all the process steps of the wind farm power generation prediction method in the above embodiments. The working principles and beneficial effects of the two correspond one by one, and thus will not be elaborated here.

[0164] Embodiments of the present invention also provide an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a wind farm power generation prediction program. When the processor executes the computer program, the steps in the above embodiments of each wind farm power generation prediction method are implemented, such as Figure 1 Step S11 shown in the figure. Alternatively, when the processor executes the computer program, the functions of each module / unit in the above system embodiments are implemented, such as the historical meteorological factor selection module.

[0165] Exemplarily, the computer program may be divided into one or more modules / units. The one or more modules / units are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the electronic device.

[0166] The electronic device may be a computing device such as a desktop computer, a notebook, a palm computer, and a smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above components are only examples of the electronic device, and do not constitute a limitation on the electronic device. It may include more or fewer components than the above, or combine some components, or different components. For example, the electronic device may further include input / output devices, network access devices, a bus, etc.

[0167] The so-called processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the electronic device and connects all parts of the entire electronic device through various interfaces and lines.

[0168] The memory can be used to store the computer programs and / or modules. The processor realizes various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory and by calling the data stored in the memory. The memory mainly includes a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created according to the use of the mobile phone (such as audio data, phone book, etc.). In addition, the memory may include high-speed random access memory and may also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage devices.

[0169] Among them, if the modules / units integrated in the electronic device are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above-described embodiment methods of the present invention, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or system, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0170] It should be noted that the system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the accompanying drawings of the system embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which can be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art can understand and implement it without creative effort.

[0171] The specific embodiments described above have further elaborated on the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only for the specific embodiments of the present invention and is not used to limit the protection scope of the present invention. It is particularly pointed out that for those skilled in the art, any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for predicting wind farm power generation, characterized in that: include: Obtain historical operation data and historical meteorological data of the wind farm, as well as recent data of the wind farm, and establish an initial power generation prediction model; the recent data includes recent meteorological data, recent operation data and recent output log data within a preset time period from the current moment; wherein the initial power generation prediction model includes a power prediction model based on physical factors and a statistical model based on a statistical analysis method, the power prediction model includes a power function of the generated power with respect to time and wind speed, and the statistical model includes a statistical function of the wind farm unit output with respect to wind speed; Selecting historical meteorological factors related to wind farm power generation from the historical meteorological data; Selecting historical power generation related to the historical meteorological factors from the historical operation data; The initial power generation prediction model is trained using the historical meteorological factors and the historical power generation to obtain a first power generation prediction model, including: Using a polynomial fitting method, the power prediction model is trained according to the historical power generation to obtain a target power prediction model; Using the least square method, the statistical function is calculated according to the historical meteorological factors, and the statistical function is substituted into the statistical model to obtain a target statistical model; Performing a model fusion operation on the target power prediction model and the target statistical model to obtain the first power generation prediction model; The statistical function is calculated according to the following formula: in, Indicated in The statistical function of time, Indicated in Wind speed at the moment Second power, Indicates the average wind speed during the period of the historical meteorological factors. Indicates the number of all time periods recorded in the historical meteorological factor; Performing multi-scale feature extraction on the recent data to obtain multi-scale feature data, and inputting the multi-scale feature data into a multi-scale normalized flow model to obtain multi-scale normalized flow data; Inputting the multi-scale normalized flow data into a hypersphere support vector machine model to perform multi-dimensional time series data classification to obtain classified output time series data; Calculate the contribution score according to the output time series data to obtain a real-time contribution score for each output time series data; When it is determined that the real-time contribution score is greater than a preset score threshold, the first power generation prediction model is trained using the recent data to obtain a second power generation prediction model; when it is determined that the real-time contribution score is less than or equal to the preset score threshold, the first power generation prediction model is directly output as the second power generation prediction model; The second power generation prediction model is used to predict the future power generation of the wind farm to obtain the power generation data of the wind farm.

2. The method for predicting wind farm power generation according to claim 1, characterized in that: The multi-scale normalized flow data is input into the hypersphere support vector machine model to perform multi-dimensional time series data classification to obtain the classified output time series data, including: According to the multi-scale normalized flow data, a hypersphere is constructed in a feature space to obtain hypersphere data; According to the multi-scale normalized flow data and the hypersphere data, a data projection operation is performed and the distance between the data point and the center of the sphere is calculated to obtain the center-of-sphere distance data; According to the ball center distance data, multi-dimensional time series data classification is performed to obtain classified output time series data.

3. The method for predicting wind farm power generation according to claim 2, characterized in that: The step of calculating the contribution score according to the output time series data to obtain the real-time contribution score of each output time series data includes: According to the output time series data, the mean value of the core time series data is calculated to obtain the core center point; According to the output time series data and the core center point, the Euclidean distance between each time series data and the core center point is calculated to obtain a close distance; The contribution score of the time series data is calculated according to the proximity distance to obtain a real-time contribution score of each time series data.

4. The method for predicting wind farm power generation according to claim 1, characterized in that: The historical operation data includes the output and power of the wind farm units, and the corresponding power generation; the historical meteorological data includes the historical wind speed and the corresponding wind direction.

5. The method for predicting wind farm power generation according to claim 1, characterized in that: The selecting of historical meteorological factors related to the power generation of the wind farm from the historical meteorological data includes: Setting a correlation coefficient threshold between the historical meteorological data and the wind farm power generation; Calculating the correlation coefficient between the historical meteorological data and the power generation of the wind farm; The historical meteorological data corresponding to the correlation coefficient being greater than the correlation coefficient threshold is selected and output as the historical meteorological factor.

6. The method for predicting wind farm power generation according to claim 5, characterized in that: The calculating of the correlation coefficient between the historical meteorological data and the power generation of the wind farm comprises: The correlation coefficient is calculated according to the following formula: in, represents the correlation coefficient, Indicated in The wind speed at the moment, Indicates the average wind speed during the period of historical meteorological data. Indicated in The wind farm power generation at the time, represents the average value of wind farm power generation during the period of historical meteorological data, Represents the number of all time periods recorded in the historical meteorological factor.

7. The method for predicting wind farm power generation according to claim 1, characterized in that: The method of using a polynomial fitting method to train the power prediction model according to the historical power generation to obtain a target power prediction model includes: Calculate the power function according to the historical power generation; Substituting the power function into the power prediction model to obtain the target power prediction model; The power function is calculated according to the following formula: in, Indicated in The power function at time, Indicated in The historical power generation at the time, Indicated in Wind speed at the moment.

8. A wind farm power generation prediction system, characterized in that: A method for predicting wind farm power generation according to any one of claims 1 to 7, comprising: The data acquisition module is used to obtain the historical operation data and historical meteorological data of the wind farm, as well as the recent data of the wind farm, and to establish an initial power generation prediction model; A historical meteorological factor selection module, used to select historical meteorological factors related to the power generation of the wind farm from the historical meteorological data; A historical power generation selection module, used to select historical power generation related to the historical meteorological factors from the historical operation data; A model training module, used to train the initial power generation prediction model using the historical meteorological factors and the historical power generation to obtain a first power generation prediction model; A model optimization module, used for performing multi-scale feature extraction on the recent data to obtain multi-scale feature data, and inputting the multi-scale feature data into a multi-scale normalized flow model to obtain multi-scale normalized flow data; inputting the multi-scale normalized flow data into a hypersphere support vector machine model to perform multi-dimensional time series data classification to obtain classified output time series data; performing contribution score calculation based on the output time series data to obtain a real-time contribution score for each output time series data; when it is determined that the real-time contribution score is greater than a preset score threshold, using the recent data to train the first power generation prediction model to obtain a second power generation prediction model; when it is determined that the real-time contribution score is less than or equal to the preset score threshold, directly outputting the first power generation prediction model as the second power generation prediction model; The model prediction module is used to predict the future power generation of the wind farm using the second power generation prediction model to obtain the power generation data of the wind farm.

Images