Generated power prediction method and device, storage medium and electronic device

Abstract

The invention discloses a generation power prediction method and device, a storage medium and an electronic device, and the method comprises the steps: obtaining typhoon data corresponding to a typhoon environment under the condition that a wind power plant is determined to be in the typhoon environment, and the typhoon data at least comprises the wind speed of the typhoon in the typhoon environment; the generation power of the wind power plant is predicted based on the typhoon data through a wind power prediction model, and the wind power prediction model is a model obtained through fusion data training corresponding to multi-source meteorological data. The multi-source meteorological data comprises historical typhoon data respectively collected by a plurality of data sources in a historical typhoon environment. The problem that a traditional wind power prediction model is poor in prediction accuracy in a typhoon environment is solved.

Description

Technical Field

[0001] The present application relates to the technical field of wind power generation, and in particular to a method and device for predicting power generation, a storage medium and an electronic device. Background Art

[0002] In areas rich in wind energy resources, especially in typhoon-prone areas, wind farms face unique challenges and opportunities. Typhoons, as an extreme meteorological phenomenon, can significantly improve the power generation efficiency of wind turbines under certain conditions due to the strong winds they bring. Especially when the typhoon intensity is moderate and the wind speed falls within the designed full-power wind speed range of the wind turbine (11m / s to 25m / s), the increase in power generation makes an important contribution to the annual output of the wind farm. However, the unpredictability of typhoon paths, the rapid changes in wind speed and direction information, and the significant fluctuations in the power output characteristics of wind farms under extreme weather conditions all pose huge challenges to wind power forecasting.

[0003] Traditional wind power prediction models are often not fully considered in their design for extreme weather conditions such as typhoons, and have the problem of poor prediction accuracy. Traditional wind power prediction models mainly rely on conventional meteorological data, such as wind speed, wind direction, temperature, etc., but lack the ability to analyze and predict the typhoon path, intensity changes, and the resulting rapid fluctuations in wind speed and wind direction. This limitation makes it difficult for the model to capture instantaneous changes in wind speed and direction in a typhoon environment, thus affecting the accuracy of the prediction. Inaccurate power prediction will not only increase the "two detailed rules" assessment costs of wind farms due to their inability to accurately respond to grid dispatching needs, but also cause additional power generation losses and operating costs when wind turbines implement emergency shutdown strategies when wind speeds exceed the design safety threshold (25m / s).

[0004] In the related technology, there is no effective solution to the problem that the traditional wind power prediction model has poor prediction accuracy in typhoon environment. Summary of the invention

[0005] The embodiments of the present application provide a method and device for predicting power generation, a storage medium, and an electronic device to at least solve the problem of poor prediction accuracy of traditional wind power prediction models in typhoon environments in the prior art.

[0006] According to an embodiment of the present application, a method for predicting power generation is provided, comprising: when it is determined that a wind farm is in a typhoon environment, obtaining typhoon data corresponding to the typhoon environment, wherein the typhoon data at least includes: the wind speed of the typhoon in the typhoon environment; predicting the power generation of the wind farm based on the typhoon data using a wind power prediction model, wherein the wind power prediction model is a model obtained by training with fusion data corresponding to multi-source meteorological data, and the multi-source meteorological data includes: historical typhoon data collected separately from multiple data sources in historical typhoon environments.

[0007] In an exemplary embodiment, before predicting the power generation power of the wind farm based on the typhoon data through a wind power prediction model, the method also includes: fusing the multi-source meteorological data through a data assimilation algorithm to obtain the fused data; and extracting a typhoon influence factor from the multi-source meteorological data, wherein the typhoon influence factor includes at least: typhoon position, cloud diameter, typhoon central pressure, and typhoon turbulence intensity; and training a deep learning model through the fused data and the typhoon influence factor to obtain the wind power prediction model.

[0008] In an exemplary embodiment, the multi-source meteorological data is fused through a data assimilation algorithm to obtain the fused data, including: acquiring the multi-source meteorological data under the historical typhoon environment collected by the multiple data sources, wherein the multi-source meteorological data at least includes: meteorological data collected by ground observation stations, meteorological radar data, satellite observation data, and numerical weather forecast data; performing a preprocessing operation on the multi-source meteorological data, wherein the preprocessing operation at least includes: a data noise reduction operation and a data format unification operation; and fusing the multi-source meteorological data after the preprocessing operation through the data assimilation algorithm to obtain the fused data.

[0009] In an exemplary embodiment, extracting typhoon impact factors from the multi-source meteorological data includes: determining the cloud system diameter through typhoon cloud map photos collected by meteorological satellites, wherein the multi-source meteorological data includes: the typhoon cloud map photos; and analyzing the typhoon position through the typhoon cloud map photos at different times, wherein the typhoon position includes: typhoon distance, moving speed, and moving direction.

[0010] In an exemplary embodiment, extracting typhoon impact factors from the multi-source meteorological data includes: determining the typhoon central pressure through meteorological satellite remote sensing data in the multi-source meteorological data; and determining the typhoon turbulence intensity through data collected by monitoring equipment in the multi-source meteorological data.

[0011] In an exemplary embodiment, the wind power prediction model is obtained by training a deep learning model with the fused data and the typhoon impact factor, including: taking the typhoon impact factor as an input feature of the deep learning model; and obtaining the wind power prediction model by training a deep learning model with the input feature with the fused data.

[0012] According to another embodiment of the embodiments of the present application, a device for predicting power generation is also provided, including: an acquisition module, used to obtain typhoon data corresponding to the typhoon environment when it is determined that the wind farm is in a typhoon environment, wherein the typhoon data at least includes: the wind speed of the typhoon in the typhoon environment; a prediction module, used to predict the power generation of the wind farm based on the typhoon data through a wind power prediction model, wherein the wind power prediction model is a model obtained by training with fusion data corresponding to multi-source meteorological data, and the multi-source meteorological data includes: historical typhoon data collected separately by multiple data sources in historical typhoon environments.

[0013] According to another aspect of the embodiments of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the above method when running.

[0014] According to another aspect of the embodiments of the present application, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the method through the computer program.

[0015] According to another aspect of the embodiments of the present application, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0016] In an embodiment of the present application, when it is determined that the wind farm is in a typhoon environment, typhoon data corresponding to the typhoon environment is obtained, wherein the typhoon data at least includes: the wind speed of the typhoon in the typhoon environment; the wind power prediction model is used to predict the power generation of the wind farm based on the typhoon data, wherein the wind power prediction model is a model obtained by training fusion data corresponding to multi-source meteorological data, and the multi-source meteorological data includes: historical typhoon data collected by multiple data sources in historical typhoon environments. Through the above embodiments, the problem of poor prediction accuracy of traditional wind power prediction models in typhoon environments in the prior art is solved, and the prediction accuracy of the power generation of wind farms is improved, so that the wind power prediction model can be applied to both conventional wind resource environments and typhoon environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0018] Figure 1 It is a hardware structure block diagram of a computer terminal of a method for predicting power generation according to an embodiment of the present application;

[0019] Figure 2 is a flow chart of a method for predicting power generation according to an embodiment of the present application;

[0020] Figure 3 is a schematic diagram of a processing process of a method for predicting power generation according to an embodiment of the present application;

[0021] Figure 4 It is a structural block diagram of a device for predicting power generation according to an embodiment of the present application. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0023] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices; "plurality" means two or more.

[0024] The method embodiments provided in the embodiments of the present application can be executed in a computer terminal or similar computing device or cloud platform or independent physical server or software platform, wherein the above software platform is run through one or more servers. Taking running on a computer terminal as an example, Figure 1FIG. 1 is a hardware structure block diagram of a computer terminal for a method for predicting power generation according to an embodiment of the present application. Figure 1 As shown, the computer terminal may include one or more ( Figure 1 Only one is shown in the figure) processor 102 and memory 104 for storing data. In an exemplary embodiment, the computer terminal may also include a transmission device 106 for communication functions and an input and output device 108, wherein the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA. A person skilled in the art can understand that Figure 1 The structure shown is only for illustration and does not limit the structure of the above-mentioned computer terminal. Figure 1 More or fewer components as shown, or with Figure 1 Equivalent functions or comparisons shown Figure 1 A different configuration with more features is shown.

[0025] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as computer programs corresponding to the methods in the embodiments of the present application. The processor 102 executes various functional applications and data processing by running the computer programs stored in the memory 104, that is, the above-mentioned methods are implemented. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the computer terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0026] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of a computer terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0027] In this embodiment, a method for predicting power generation is provided, which is applied to the above-mentioned computer terminal. Figure 2 : is a flow chart of a method for predicting power generation according to an embodiment of the present application, the process includes the following steps:

[0028] Step S202, when it is determined that the wind farm is in a typhoon environment, obtaining typhoon data corresponding to the typhoon environment, wherein the typhoon data at least includes: a wind speed of the typhoon in the typhoon environment;

[0029] Step S204, predicting the power generation of the wind farm based on the typhoon data through a wind power prediction model, wherein the wind power prediction model is a model trained by fusion data corresponding to multi-source meteorological data, and the multi-source meteorological data includes: historical typhoon data collected by multiple data sources in historical typhoon environments.

[0030] Through the above steps, when it is determined that the wind farm is in a typhoon environment, typhoon data corresponding to the typhoon environment are obtained, wherein the typhoon data at least includes: the wind speed of the typhoon in the typhoon environment; the wind power prediction model is used to predict the power generation of the wind farm based on the typhoon data, wherein the wind power prediction model is a model obtained by training the fusion data corresponding to multi-source meteorological data, and the multi-source meteorological data includes: historical typhoon data collected by multiple data sources in historical typhoon environments. Through the above embodiments, the problem of poor prediction accuracy of traditional wind power prediction models in typhoon environments in the prior art is solved, and the prediction accuracy of the power generation of wind farms is improved, so that the wind power prediction model can be applied to both conventional wind resource environments and typhoon environments.

[0031] In an exemplary embodiment, before predicting the power generation power of the wind farm based on the typhoon data through a wind power prediction model, the method also includes: fusing the multi-source meteorological data through a data assimilation algorithm to obtain the fused data; and extracting a typhoon influence factor from the multi-source meteorological data, wherein the typhoon influence factor includes at least: typhoon position, cloud diameter, typhoon central pressure, and typhoon turbulence intensity; and training a deep learning model through the fused data and the typhoon influence factor to obtain the wind power prediction model.

[0032] In the embodiment of the present application, the deep learning model is trained by combining data assimilation technology and feature extraction technology to obtain a wind power prediction model. Specifically:

[0033] First, it is necessary to integrate multi-source meteorological data from multiple data sources to ensure that the wind power prediction model can obtain the most accurate meteorological information (i.e., fused data). Data assimilation algorithms, such as Kalman filtering or variational methods, can handle the noise and uncertainty in multi-source meteorological data. Through continuous adjustment and optimization, the estimation of the model state is closer to the actual observation, thereby obtaining high-precision fused data.

[0034] Secondly, in order to more accurately assess the impact of typhoons on wind farm power generation, it is also necessary to extract specific typhoon-related features, namely typhoon impact factors, from multi-source meteorological data. These typhoon impact factors include but are not limited to: typhoon location; cloud diameter reflecting the scale of the typhoon; typhoon center pressure reflecting the intensity of the typhoon; typhoon turbulence intensity. Among them, the typhoon location can further include: the typhoon distance between the typhoon center and the wind farm, as well as the typhoon's moving speed and direction.

[0035] Finally, the fused data and the extracted typhoon impact factors are used as input features for training deep learning models. The deep learning model can be a model based on the Transformer architecture, which can process complex time series data and capture the rapid changes of meteorological elements such as wind speed and wind direction during the typhoon process through its unique attention mechanism. In addition, the residual learning and gating mechanisms in the model can effectively solve the nonlinear and non-stationary problems of data under typhoon conditions and improve the prediction accuracy of the model. Finally, the trained deep learning model, namely the wind power prediction model, can better adapt to extreme weather conditions such as typhoons, improve the accuracy of predictions by considering the influence of typhoon-specific elements, and provide more reliable data support for wind farm operations and power grid dispatching.

[0036] In some optional embodiments, the multi-source meteorological data are fused through a data assimilation algorithm to obtain the fused data, including: acquiring the multi-source meteorological data under the historical typhoon environment collected by the multiple data sources, wherein the multi-source meteorological data at least includes: meteorological data collected by ground observation stations, meteorological radar data, satellite observation data, and numerical weather forecast data; performing a preprocessing operation on the multi-source meteorological data, wherein the preprocessing operation at least includes: a data noise reduction operation and a data format unification operation; and fusing the multi-source meteorological data after the preprocessing operation through the data assimilation algorithm to obtain the fused data.

[0037] It can be understood that the multiple data sources include the above-mentioned ground observation stations, meteorological radars that collect meteorological radar data, meteorological satellites that collect satellite observation data, equipment that collects the numerical weather forecast data, etc.

[0038] Each source of meteorological data in multi-source meteorological data has its own unique advantages and limitations. For example, meteorological data collected by ground observation stations can provide local and accurate data such as wind speed, wind direction, humidity, and temperature; meteorological radar data can reveal the three-dimensional structure and precipitation distribution of wind fields (such as typhoon wind fields); satellite observation data has a wide coverage and can obtain a macroscopic view of typhoon cloud systems; numerical weather forecast data provides predictions of future meteorological conditions. By collecting these multi-source meteorological data under historical typhoon conditions, a comprehensive meteorological information dataset can be constructed for subsequent analysis and modeling.

[0039] The collected multi-source meteorological data may be inconsistent in terms of format, quality, timestamp, etc., so preprocessing operations are required to ensure the validity and consistency of multi-source meteorological data. Preprocessing operations include at least two parts: data noise reduction and data format unification: Data noise reduction operation: For random noise and outliers in the data, sliding average, smoothing filter and other technologies are used to reduce the noise level and improve data quality. This step helps to remove non-meteorological factors in the data and improve the accuracy of model prediction. Data format unification operation: Convert data from all data sources into a unified format and align timestamps to facilitate the integration of these data into the same model for analysis. This step is the prerequisite for data fusion and ensures the comparability of data and the feasibility of model training.

[0040] Furthermore, the pre-processed multi-source meteorological data is fused through the data assimilation algorithm to obtain fused data: Data assimilation algorithms, such as Kalman filtering and variational methods, can be used to integrate the pre-processed multi-source meteorological data to generate fused data. These algorithms can take into account the differences in temporal and spatial resolution of different data sources, as well as the uncertainty and credibility of the data, and optimize the multi-source meteorological data through mathematical methods, so that the prediction model can use the most accurate and comprehensive meteorological information. The obtained fused data not only contains the advantages of the meteorological data of each data source, but also makes up for the limitations of a single data source, providing a high-quality data foundation for subsequent model training.

[0041] At the same time, before or after acquiring the fused data, extracting typhoon impact factors from the multi-source meteorological data, including: determining the cloud system diameter through typhoon cloud map photos collected by meteorological satellites, wherein the multi-source meteorological data include: the typhoon cloud map photos; and analyzing the typhoon position through the typhoon cloud map photos at different times, wherein the typhoon position includes: typhoon distance, moving speed, and moving direction.

[0042] It is understandable that in the process of determining the cloud diameter through typhoon cloud photos collected by meteorological satellites, the typhoon cloud photos collected by meteorological satellites include: visible light typhoon cloud photos, infrared typhoon cloud photos, and the typhoon position and cloud diameter can be determined through the typhoon cloud photos. More specifically, determining the typhoon position includes: determining the typhoon center position at different times through typhoon cloud photos at different times, and calculating the typhoon distance, moving speed and moving direction of the typhoon through the typhoon center position at different times.

[0043] Optionally, extracting typhoon impact factors from the multi-source meteorological data also includes: determining the typhoon central pressure through meteorological satellite remote sensing data in the multi-source meteorological data; and determining the typhoon turbulence intensity through data collected by monitoring equipment in the multi-source meteorological data.

[0044] Specifically, through meteorological satellite remote sensing data, the electromagnetic radiation intensity of different wavelength channels in the typhoon area is analyzed to obtain the central air pressure of the typhoon.

[0045] Specifically, Doppler radar and other monitoring equipment are used to obtain the precipitation distribution and wind field structure inside the typhoon, and the wind speed inside the typhoon is analyzed by methods such as standard deviation method and turbulent kinetic energy method to obtain the typhoon turbulence intensity. Optionally, the multiple data sources also include: the above-mentioned monitoring equipment. Multi-source meteorological data also includes: data collected by monitoring equipment, such as precipitation distribution and wind field structure inside the typhoon.

[0046] In an exemplary embodiment, the wind power prediction model is obtained by training a deep learning model with the fused data and the typhoon impact factor, including: taking the typhoon impact factor as an input feature of the deep learning model; and obtaining the wind power prediction model by training a deep learning model with the input feature with the fused data.

[0047] Optionally, factors such as wind speed, wind direction, temperature, humidity, and pressure at the wind farm location are used as input features of the deep learning model, and the extracted typhoon impact factors are also used as input features of the deep learning model. The deep learning model is trained using the power generation of the wind farm as the output feature of the deep learning model, and finally a wind power prediction model is obtained.

[0048] In order to better understand the process of the above-mentioned power generation prediction method, the above-mentioned power generation prediction method is described below in combination with an optional embodiment, but is not used to limit the technical solution of the embodiment of the present application.

[0049] The solution provided in the optional embodiment of the present application is intended to improve the power prediction accuracy of wind farms under extreme weather conditions such as typhoons and enhance the operational stability and safety of wind power systems. Figure 3 As shown, by integrating meteorological data assimilation technology, typhoon impact factor extraction technology, and deep learning algorithm optimization method, the optional embodiments of the present application can more accurately capture the complex impact of typhoons on wind power output, and provide a reliable basis for wind farm scheduling, power grid operation planning and energy management.

[0050] The optional embodiment of the present application can be implemented by the following steps:

[0051] Step 1: Fusion and assimilation of multi-source meteorological data.

[0052] Traditional wind power prediction models mostly rely on a single meteorological data source, but the meteorological elements in a typhoon process change dramatically, and a single meteorological source data cannot accurately predict the changes in wind speed, wind direction and other elements brought by the typhoon. The optional embodiment of this application introduces multi-source meteorological data, and efficiently integrates it through advanced data assimilation algorithms to improve the prediction accuracy of key parameters such as typhoon path, intensity, wind speed and wind direction. The fusion and assimilation of multi-source meteorological data for typhoon processes mainly includes the following steps:

[0053] Step 11: Multi-source meteorological data collection. The multi-source meteorological data used in the optional embodiment of the present application includes data from ground observation stations, meteorological radar data, satellite observation data, numerical weather forecast data, etc. Meteorological elements include wind speed, wind direction, temperature, air pressure, humidity, satellite cloud images, typhoon path, typhoon intensity, etc.

[0054] Step 12: Preprocessing of multi-source meteorological data. Raw meteorological data from different sources have problems such as different data formats and poor data quality, so data preprocessing is required, mainly including:

[0055] (1) Outlier elimination: Use the 3 sigma method, k-nearest neighbor algorithm, etc. to identify and eliminate extreme outliers in the data set (i.e., multi-source meteorological data).

[0056] (2) Data interpolation: For missing data points and outlier points that have been removed, linear interpolation, cubic spline interpolation, Newton interpolation and other methods are used to supplement them.

[0057] (3) Data denoising: Random noise in time series data (i.e., multi-source meteorological data) is removed through sliding average, smoothing filter and other techniques to reduce the interference of high-frequency noise data on model prediction.

[0058] (4) Data format unification: All time series data are converted into a unified format and timestamps are aligned to facilitate subsequent processing.

[0059] Step 13: Fusion and assimilation of multi-source meteorological data.

[0060] Meteorological data from different sources have different spatiotemporal resolutions. Data assimilation and fusion technology can fuse meteorological observation data with multi-source meteorological data such as numerical weather forecast data into a more accurate meteorological data set (i.e., fused data) with consistent spatiotemporal resolution. Specifically, the optional embodiments of the present application use Kalman filtering, variational methods, etc. to fuse and assimilate multi-source meteorological data to obtain a data set for model training. Among them, Kalman filtering is a method based on probability statistics, which realizes data assimilation by continuously updating the estimated value of the model state. The variational method is to construct an objective function and optimize the difference between the observed data and the model prediction by mathematical methods to minimize the difference, thereby obtaining the optimal model.

[0061] Step 14: Data evaluation and optimization. The assimilated data (i.e., fused data) is compared with independent observation data or historical data to verify its accuracy and reliability. If the assimilation result is not accurate enough, it is optimized by adjusting the assimilation method, optimizing model parameters, and improving the data preprocessing process.

[0062] Through the above steps, a high-precision weather forecast data set (i.e., fused data) within the wind farm range can be obtained.

[0063] Step 2: Extraction of typhoon impact factors.

[0064] Traditional wind power prediction models only use meteorological factors such as wind speed, wind direction, and temperature. In order to consider the impact of typhoons on power prediction, the optional embodiment of the present application extracts typhoon impact factors from multi-source meteorological data and uses them as input variables of the wind power prediction model, thereby improving the accuracy of wind power prediction under the influence of typhoons. The steps of extracting typhoon impact factors include:

[0065] The visible light and infrared typhoon cloud images were taken by meteorological satellites to obtain the typhoon position and cloud diameter, and further calculate the distance between the typhoon position and the wind farm location. The typhoon moving speed and direction were calculated by analyzing the typhoon center position at different times. The typhoon distance, moving speed, moving direction, and cloud diameter were used as four typhoon influencing factors.

[0066] Using meteorological satellite remote sensing data, the electromagnetic radiation intensity of different wavelength channels in the typhoon area is analyzed to obtain the typhoon center pressure as a typhoon influencing factor.

[0067] Doppler radar and other monitoring equipment are used to obtain the precipitation distribution and wind field structure inside the typhoon. The wind speed inside the typhoon is analyzed through the standard deviation method, turbulent kinetic energy method and other methods to obtain the typhoon turbulence intensity as a typhoon influencing factor.

[0068] Through the above method, a total of 6 time series data of typhoon influencing factors are obtained, which are aligned in time with the meteorological forecast data set (i.e., fused data) obtained in step 1.

[0069] Step 3: Deep learning algorithm training.

[0070] Combining high-precision meteorological data and typhoon impact factor data, the wind farm power generation is predicted based on the deep learning algorithm (i.e. deep learning model). Including:

[0071] Model training: In the accurate meteorological data (i.e. fused data) obtained after melting multi-source meteorological data, factors such as wind speed, wind direction, temperature, humidity, and pressure at the wind farm location are used as input features of the deep learning algorithm. At the same time, the typhoon distance, moving speed, moving direction, cloud diameter, typhoon center pressure, and typhoon turbulence intensity, a total of 6 typhoon influencing factors, are also used as input features of the deep learning algorithm. The wind farm power generation is used as the output to train the deep learning model. In terms of algorithms, deep neural networks (such as the Transformer model) are used to predict wind power, and the attention mechanism is combined to enhance the model's ability to capture important time points in time series data (such as the turning point of rapid intensification and attenuation of typhoons). In view of the nonlinear and non-stationary characteristics during typhoons, the learning ability and generalization ability of the model are improved by introducing technical means such as residual learning and gating mechanisms to ensure the prediction accuracy under extreme weather conditions.

[0072] Dynamic adjustment of prediction parameters: According to the typhoon warning level, historical typhoon impact data and real-time meteorological forecast results, the internal parameters of the wind power prediction model, such as the number of network layers and the number of neurons, are dynamically adjusted to adapt to the different impacts of typhoons of different intensities and paths on wind power, and to achieve more refined predictions.

[0073] This completes the establishment of a wind power prediction model that takes into account the impact of typhoons.

[0074] To summarize, the optional embodiments of the present application obtain high-precision meteorological forecast data through meteorological data assimilation technology, and consider the impact of typhoon processes on the power generation of wind farms by extracting typhoon influencing factors. The two parts of data are used as input features for training the power prediction model. Through deep learning algorithms and dynamic adjustment of prediction parameters, refined modeling and typhoon adaptability of the wind power prediction model are achieved, thereby improving the prediction accuracy.

[0075] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of each embodiment of the present application.

[0076] The present application also provides a structural block diagram of a power generation prediction device. Figure 4 is a structural block diagram of a device for predicting power generation according to an embodiment of the present application; Figure 4 As shown, including:

[0077] The acquisition module 42 is used to acquire typhoon data corresponding to the typhoon environment when it is determined that the wind farm is in a typhoon environment, wherein the typhoon data at least includes: the wind speed of the typhoon in the typhoon environment;

[0078] The prediction module 44 is used to predict the power generation of the wind farm based on the typhoon data through a wind power prediction model, wherein the wind power prediction model is a model obtained by training the fusion data corresponding to multi-source meteorological data, and the multi-source meteorological data includes: historical typhoon data collected by multiple data sources in historical typhoon environments.

[0079] Through the above device, when it is determined that the wind farm is in a typhoon environment, typhoon data corresponding to the typhoon environment are obtained, wherein the typhoon data at least includes: the wind speed of the typhoon in the typhoon environment; the wind power prediction model is used to predict the power generation of the wind farm based on the typhoon data, wherein the wind power prediction model is a model obtained by training the fusion data corresponding to multi-source meteorological data, and the multi-source meteorological data includes: historical typhoon data collected by multiple data sources in historical typhoon environments. Through the above embodiment, the problem of poor prediction accuracy of the traditional wind power prediction model in the prior art in a typhoon environment is solved, and the prediction accuracy of the power generation of the wind farm is improved, so that the wind power prediction model can be applied to both conventional wind resource environments and typhoon environments.

[0080] In an exemplary embodiment, the device also includes a training module, which is used for: before predicting the power generation power of the wind farm based on the typhoon data through a wind power prediction model: fusing the multi-source meteorological data through a data assimilation algorithm to obtain the fused data; and extracting typhoon influence factors from the multi-source meteorological data, wherein the typhoon influence factors include at least: typhoon position, cloud system diameter, typhoon central pressure, typhoon turbulence intensity; training a deep learning model through the fused data and the typhoon influence factors to obtain the wind power prediction model.

[0081] In an exemplary embodiment, the training module is also used to: obtain the multi-source meteorological data under the historical typhoon environment collected by the multiple data sources, wherein the multi-source meteorological data at least includes: meteorological data collected by ground observation stations, meteorological radar data, satellite observation data, and numerical weather forecast data; perform preprocessing operations on the multi-source meteorological data, wherein the preprocessing operations at least include: data noise reduction operations, data format unification operations; and fuse the multi-source meteorological data after the preprocessing operations through the data assimilation algorithm to obtain the fused data.

[0082] In an exemplary embodiment, the training module is also used to: determine the cloud system diameter through typhoon cloud map photos collected by meteorological satellites, wherein the multi-source meteorological data includes: the typhoon cloud map photos; and analyze the typhoon position through the typhoon cloud map photos at different times, wherein the typhoon position includes: typhoon distance, moving speed, and moving direction.

[0083] In an exemplary embodiment, the training module is also used to: determine the central air pressure of the typhoon through meteorological satellite remote sensing data in the multi-source meteorological data; and determine the typhoon turbulence intensity through data collected by monitoring equipment in the multi-source meteorological data.

[0084] In an exemplary embodiment, the training module is further used to: use the typhoon impact factor as an input feature of the deep learning model; and train the deep learning model with the input feature through the fusion data to obtain the wind power prediction model.

[0085] An embodiment of the present application further provides a storage medium, which includes a stored program, wherein the program executes any of the above methods when it is run.

[0086] Optionally, in this embodiment, the storage medium may be configured to store program codes for executing the following steps:

[0087] S1, when it is determined that the wind farm is in a typhoon environment, obtaining typhoon data corresponding to the typhoon environment, wherein the typhoon data at least includes: a wind speed of the typhoon in the typhoon environment;

[0088] S2, predicting the power generation of the wind farm based on the typhoon data through a wind power prediction model, wherein the wind power prediction model is a model trained by fusion data corresponding to multi-source meteorological data, and the multi-source meteorological data includes: historical typhoon data collected by multiple data sources in historical typhoon environments.

[0089] An embodiment of the present application further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0090] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0091] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0092] S1, when it is determined that the wind farm is in a typhoon environment, obtaining typhoon data corresponding to the typhoon environment, wherein the typhoon data at least includes: a wind speed of the typhoon in the typhoon environment;

[0093] S2, predicting the power generation of the wind farm based on the typhoon data through a wind power prediction model, wherein the wind power prediction model is a model trained by fusion data corresponding to multi-source meteorological data, and the multi-source meteorological data includes: historical typhoon data collected by multiple data sources in historical typhoon environments.

[0094] Optionally, in this embodiment, the above-mentioned storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store program codes.

[0095] An embodiment of the present application further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps in any one of the above method embodiments are implemented.

[0096] An embodiment of the present application further provides another computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any of the above method embodiments are implemented.

[0097] An embodiment of the present application also provides a computer program, which includes computer instructions stored in a computer-readable storage medium; a processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps in any one of the above method embodiments.

[0098] Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments and optional implementation modes, and this embodiment will not be described in detail here.

[0099] Obviously, those skilled in the art should understand that the above modules or steps of the present application can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order from that herein, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0100] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for predicting power generation, characterized in that: include: When it is determined that the wind farm is in a typhoon environment, typhoon data corresponding to the typhoon environment are obtained, wherein the typhoon data at least include: the wind speed of the typhoon in the typhoon environment; and the generated power of the wind farm is predicted based on the typhoon data by using a wind power prediction model, wherein the wind power prediction model is a model obtained by training with fusion data corresponding to multi-source meteorological data, and the multi-source meteorological data include: historical typhoon data collected by multiple data sources in historical typhoon environments.

2. The method for predicting power generation according to claim 1, characterized in that: Before predicting the power generation of the wind farm based on the typhoon data using a wind power prediction model, the method further includes: fusing the multi-source meteorological data through a data assimilation algorithm to obtain the fused data; and Extracting typhoon impact factors from the multi-source meteorological data, wherein the typhoon impact factors at least include: typhoon position, cloud diameter, typhoon center pressure, and typhoon turbulence intensity; The wind power prediction model is obtained by training a deep learning model with the fusion data and the typhoon impact factor.

3. The method for predicting power generation according to claim 2, characterized in that: The multi-source meteorological data are fused by a data assimilation algorithm to obtain the fused data, including: Acquire the multi-source meteorological data in the historical typhoon environment collected by the multiple data sources, wherein the multi-source meteorological data at least includes: meteorological data collected by ground observation stations, meteorological radar data, satellite observation data, and numerical weather forecast data; Performing a preprocessing operation on the multi-source meteorological data, wherein the preprocessing operation at least includes: a data noise reduction operation and a data format unification operation; The multi-source meteorological data after preprocessing operation is fused through the data assimilation algorithm to obtain the fused data.

4. The method for predicting power generation according to claim 2, characterized in that: Extracting typhoon impact factors from the multi-source meteorological data includes: The cloud system diameter is determined by using typhoon cloud picture photos collected by meteorological satellites, wherein the multi-source meteorological data includes: the typhoon cloud picture photos; and The typhoon position is analyzed by using the typhoon cloud map photos at different times, wherein the typhoon position includes: typhoon distance, moving speed, and moving direction.

5. The method for predicting power generation according to claim 2, characterized in that: Extracting typhoon impact factors from the multi-source meteorological data includes: Determine the typhoon center pressure by using the meteorological satellite remote sensing data in the multi-source meteorological data; and The typhoon turbulence intensity is determined by data collected by monitoring equipment in the multi-source meteorological data.

6. The method for predicting power generation according to claim 2, characterized in that: The deep learning model is trained by the fusion data and the typhoon impact factor to obtain the wind power prediction model, including: Using the typhoon impact factor as an input feature of the deep learning model; The wind power prediction model is obtained by training a deep learning model with the input features through the fusion data.

7. A device for predicting power generation, characterized in that: include: An acquisition module is used to acquire typhoon data corresponding to the typhoon environment when it is determined that the wind farm is in a typhoon environment, wherein the typhoon data at least includes: a wind speed of the typhoon in the typhoon environment; A prediction module is used to predict the power generation of the wind farm based on the typhoon data through a wind power prediction model, wherein the wind power prediction model is a model obtained by training fused data corresponding to multi-source meteorological data, and the multi-source meteorological data includes: historical typhoon data collected by multiple data sources in historical typhoon environments.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein the program executes the method described in any one of claims 1 to 6 when executed.

9. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to execute the method according to any one of claims 1 to 6 through the computer program.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images