Wind power plant output power prediction method, equipment and medium

Abstract

The invention relates to a wind power plant output power prediction method, device and medium, and the method comprises the steps: A1, collecting historical climate data, new energy power generation data and real-time climate data, the new energy power generation data being wind power equipment output power; a2, key features are extracted from the historical climate data, a power prediction model is established according to the key features and the output power of the wind power equipment to achieve prediction of the output power of the wind power equipment, and the power prediction model is optimized through the real-time climate data, the prediction result of the output power of the wind power equipment is updated in real time; and A3, comparing the new prediction result with the actual output power of the wind power equipment to obtain the prediction result accuracy. According to the method, the new energy power can be accurately predicted along with the change of climate conditions.

Description

Technical Field

[0001] This invention relates to the field of new energy power prediction technology, and in particular to a method, electronic device and storage medium for predicting the output power of a wind farm. Background Technology

[0002] New energy refers to energy types that have been recently developed or are under research and development in the modern energy system. Compared with traditional fossil energy (such as coal, oil and natural gas), new energy is usually clean and renewable. However, the power generation of new energy is usually fluctuating and random, making it difficult to predict its power generation.

[0003] Chinese Patent: A method and system for predicting new energy power applicable to complex climates, publication number CN118095890A. Although it can predict new energy power, existing new energy power prediction methods fail to update and optimize the power prediction in a timely manner when changes in weather, wind speed, wind direction, or climate occur after the regional situation is predicted. As a result, the predicted power results deviate from the actual results, leading to errors in the utilization of new energy power. This results in problems such as energy loss due to insufficient energy storage equipment or uneven distribution due to excessive energy storage equipment.

[0004] In summary, existing methods for predicting new energy power output are prone to errors due to changes in meteorological conditions, and there is an urgent need to optimize the methods for predicting wind farm output power. Summary of the Invention

[0005] Therefore, the technical problem to be solved by the present invention is to overcome the problem that the prediction of new energy power in the prior art will have errors due to changes in meteorological conditions.

[0006] To address the aforementioned technical problems, this invention provides a method for predicting the output power of a wind farm, comprising:

[0007] A1: Collect historical climate data, new energy power generation data, and real-time climate data, wherein the new energy power generation data refers to the output power of wind power equipment;

[0008] A2: Key features are extracted from the historical climate data. Based on these key features and the output power of the wind turbine, a power prediction model is established to predict the output power of the wind turbine. The power prediction model is then optimized using real-time climate data to achieve real-time updates to the predicted output power of the wind turbine. Specifically, this includes:

[0009] a21: Perform preprocessing operations on the historical climate data to extract key features;

[0010] a22: Establish a power prediction model based on the extracted key features and the corresponding wind power output of the wind turbines;

[0011] a23: Analyze and process the real-time climate data, and then introduce the analyzed and processed real-time climate data into the power prediction model to obtain new prediction results;

[0012] A3: The accuracy of the prediction results is obtained by comparing the new prediction results with the actual output power of the wind power equipment.

[0013] In one embodiment of the present invention, the method for preprocessing the historical climate data in step a21 includes:

[0014] B1: Establish a matching template for the required data, and set the row and column headers for the matching template. The row headers correspond to the category names of different wind power equipment, and the column headers correspond to the category names of different time periods or different scenarios.

[0015] B2: Import the collected historical climate data into the matching template: First match the category names under the column headers, then match the corresponding row headers under the column headers. Extract the data content corresponding to the matching headers and record it in the data cell where the matching column headers are located vertically and the row headers are located horizontally.

[0016] B3: After the matching template is completed, the data recorded in the matching template is retained to filter and clean up the unmatched data.

[0017] In one embodiment of the present invention, the method for establishing a power prediction model based on the extracted key features and the corresponding wind power output in step a22 includes:

[0018] C1: Extract the wind speed data from the key features and obtain the output power data of n wind power devices in the same area corresponding to the wind speed data;

[0019] C2: Construct a power prediction model based on the wind speed data and the corresponding wind power output data.

[0020] In one embodiment of the present invention, step C2 further includes: obtaining curve data on the changes in wind speed and the average output power of n wind turbines based on the power prediction model, wherein the formula for calculating the average output power of the n wind turbines is:

[0021]

[0022] Where P1 is the predicted average output power of n wind turbines in the same area, (p1+p1+...+p n) represents the total output power of n wind turbines in the same area for each cycle, where n represents the number of wind turbines in the same area.

[0023] In one embodiment of the present invention, the method for analyzing and processing the real-time climate data in step a23 includes:

[0024] D1: Extract satellite meteorological images over each forecast area;

[0025] D2: Arrange the satellite meteorological images corresponding to each cycle in sequence to form an image dataset, and perform grayscale processing on the satellite meteorological images in the image dataset to identify the corresponding wind power equipment features and satellite cloud features. Based on the feature comparison method, obtain the changes in wind power equipment features and satellite cloud features, and calculate the changes in wind speed and wind direction to realize the analysis and processing of real-time climate data.

[0026] In one embodiment of the present invention, the feature comparison method in step D2 specifically includes:

[0027] d21: Overlay satellite meteorological images according to their chronological order, and compare the features of adjacent satellite meteorological images, including:

[0028] d22: Mark the center of the wind power equipment feature area as feature point K, and mark the center of the satellite cloud feature as feature point J. The selection of satellite cloud features should keep the wind direction consistent with the wind power receiving direction of the wind power equipment.

[0029] d23: Overlay two adjacent satellite meteorological images. After overlay, the feature point K in the two adjacent satellite meteorological images will be in an overlapping state. Then, determine whether there is a deviation in wind direction or wind speed based on the position of feature point J.

[0030] In one embodiment of the present invention, the method for determining whether the wind direction or wind speed deviates based on the position of feature point J in step d23 includes:

[0031] d231: Take the initial satellite meteorological image in the image dataset of one period as the base image. If there is a difference in the position of feature point J on the satellite meteorological image and the base image after overlay, mark the center of the wind power equipment feature area corresponding to the satellite meteorological image as feature point K', mark the center of the satellite cloud feature as feature point J', and determine whether feature point J' is located in the line segment direction between feature point K and feature point J.

[0032] d232: If feature point J' is located on the line segment direction between feature point K and feature point J, then the wind direction has not changed. Calculate the corresponding wind speed value within the period to determine whether the wind speed has changed.

[0033] d233: If feature point J' deviates from the line segment direction of feature point K and feature point J, the wind direction changes. Calculate the wind speed value along the line segment direction of feature point K and feature point J within the period to determine whether the wind speed has changed.

[0034] In one embodiment of the present invention, the method for calculating the wind speed value in steps d232 and d233, and the method for determining whether the wind speed changes, include:

[0035] The origin is set at the bottom left corner of the satellite weather map, and a coordinate system is formed with the bottom edge of the image as the X-axis and the left edge as the Y-axis.

[0036] The coordinates of feature point K and feature point J in the coordinate system are labeled as K(x1, y1) and J(x2, y2), and the coordinates of feature point K' and feature point J' are labeled as K'(x1', y1') and J'(x3, y3), where feature point K and feature point K' coincide.

[0037] The formula for calculating wind speed in step d232 is:

[0038]

[0039] Where V1 represents the wind speed value over one period. Let the length of the line segment between feature point K and feature point J be denoted as . It is represented by the line segment length between feature point K' and feature point J', m is the ratio of the distance on the image to the actual distance of the feature, and T is one period time;

[0040] The formula for calculating wind speed in step d233 is:

[0041]

[0042] Wherein, V2 represents the wind speed value within one period along the line segment between feature point K and feature point J after the wind direction changes. It represents the distance moved along the line segment direction between feature point K and feature point J, and α represents the wind direction deviation angle between the line segment direction between feature point K and feature point J.

[0043] V1 and V2 are compared with the initial wind speed. If the absolute value of the difference between V1 or V2 and the initial wind speed is within the deviation threshold, the wind speed has not changed. If the absolute value of the difference between V1 or V2 and the initial wind speed is outside the deviation threshold, the wind speed has changed, which will cause the output power of the wind turbine to change.

[0044] In one embodiment of the present invention, the method for introducing the analyzed and processed real-time climate data into the power prediction model in step a23 to obtain new prediction results includes:

[0045] E1: The wind speed data in the real-time climate data after analysis and processing is introduced into the power prediction model. The power prediction model generates a new predicted power and outputs it. If the new predicted power is greater than the rated output power of the wind power equipment, the rated output power of the wind power equipment is used as the new predicted power.

[0046] E2: The formula for calculating the new predicted power in step E1 is:

[0047]

[0048] Where P2 is the new predicted output power, C is the wind energy utilization coefficient of the wind turbine (a constant value), R is the radius of the wind turbine rotor sweep, ρ is the air density of the predicted area, and V new The wind speed value over one cycle;

[0049] The air density ρ changes with wind speed and is affected by the air pressure, temperature, and humidity of the predicted area. The calculation formula is as follows:

[0050]

[0051] Where F is the atmospheric pressure value of the predicted area, q is the temperature value of the predicted area, and f is the water vapor pressure value of the predicted area.

[0052] In one embodiment of the present invention, the method for comparing the new prediction result with the actual output power of the wind power equipment in step A3 includes:

[0053] A31: Extract the predicted power results obtained from multiple cycle power prediction models, and extract the actual output power of wind power equipment in the same cycle.

[0054] A32: The predicted power result is compared with the actual output power according to the cycle. If the predicted power result is within the predetermined deviation range of the actual output power, the predicted power result is accurate; otherwise, the predicted power result is inconsistent with the actual output power.

[0055] To solve the above-mentioned technical problems, the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the wind farm output power prediction method described above.

[0056] To address the aforementioned technical problems, the present invention provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the wind farm output power prediction method described above.

[0057] The technical solution of the present invention has the following advantages compared with the prior art:

[0058] The wind farm output power prediction method described in this invention extracts key features from historical climate data, establishes a power prediction model based on these key features and the output power of wind power equipment, and then optimizes the model by analyzing and processing real-time climate data to complete the prediction of wind power output power within a cycle. This allows for the determination of the impact of wind speed on the predicted output power, ensuring that the predicted power results are within a predetermined deviation range from the actual output power and that the predicted results are more closely aligned with the actual power, thereby improving the accuracy and sensitivity of new energy power prediction.

[0059] This invention forms an image dataset by sequentially arranging satellite meteorological image data corresponding to each cycle, and then performing grayscale processing on the data in the image dataset to identify the corresponding wind power equipment features and satellite cloud features. Based on the feature comparison method, the changes in wind power equipment features and satellite cloud features are obtained, and the changes in wind speed and wind direction are calculated accordingly. Attached Figure Description

[0060] To make the content of this invention easier to understand, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings.

[0061] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0062] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can better understand and implement the present invention. However, the embodiments described are not intended to limit the present invention.

[0063] Example 1

[0064] Reference Figure 1 As shown, this invention relates to a method for predicting the output power of a wind farm, comprising:

[0065] A1: Collect historical climate data, new energy power generation data, and real-time climate data, wherein the new energy power generation data refers to the output power of wind power equipment;

[0066] A2: Key features are extracted from the historical climate data. Based on these key features and the output power of the wind turbine, a power prediction model is established to predict the output power of the wind turbine. The power prediction model is then optimized using real-time climate data to achieve real-time updates to the predicted output power of the wind turbine. Specifically, this includes:

[0067] a21: Perform preprocessing operations on the historical climate data to extract key features;

[0068] a22: Establish a power prediction model based on the extracted key features and the corresponding wind power output of the wind turbines;

[0069] a23: Analyze and process the real-time climate data, and then introduce the analyzed and processed real-time climate data into the power prediction model to obtain new prediction results;

[0070] A3: The accuracy of the prediction results is obtained by comparing the new prediction results with the actual output power of the wind power equipment.

[0071] In this embodiment, a wind farm refers to an assembly of multiple wind power devices (wind turbine generators).

[0072] Furthermore, the method for preprocessing the historical climate data in step a21 includes:

[0073] B1: Establish a matching template for the required data, and set the row and column headers for the matching template. The row headers correspond to the category names of different wind power equipment, and the column headers correspond to the category names of different time periods or different scenarios.

[0074] B2: Import the collected historical climate data into the matching template: First match the category names under the column headers, then match the corresponding row headers under the column headers. Extract the data content corresponding to the matching headers and record it in the data cell where the matching column headers are located vertically and the row headers are located horizontally.

[0075] B3: After the matching template is completed, the data recorded in the matching template is retained to filter and clean up the unmatched data.

[0076] Furthermore, the method for establishing a power prediction model based on the extracted key features and the corresponding wind power output in step a22 includes:

[0077] C1: Extract the wind speed data from the key features and obtain the output power data of n wind power devices in the same area corresponding to the wind speed data;

[0078] C2: Construct a power prediction model based on the wind speed data and the corresponding wind power output data. It should be noted that the power prediction model in this embodiment is also related to the air density ρ in the climate data. The air density ρ changes with wind speed and is affected by the air pressure, temperature, and humidity of the prediction area. Since this embodiment focuses on the impact of wind speed on power, the influence of air density ρ is not emphasized.

[0079] Furthermore, step C2 also includes: obtaining curve data on the changes in wind speed and the average output power of n wind turbines based on the power prediction model, wherein the formula for calculating the average output power of the n wind turbines is:

[0080]

[0081] Where P1 is the predicted average output power of n wind turbines in the same area, (p1+p1+...+p n Let be the total output power of n wind turbines in the same area per cycle, where n represents the number of wind turbines in the same area. It's important to note that the graph visually illustrates the relationship between wind speed and the average output power of the n wind turbines. By observing the graph, one can quickly understand the approximate output power of the n wind turbines in the same area under different wind speeds, facilitating relevant decision-making. Furthermore, the graph can help verify the accuracy of the power prediction model: if the output power predicted by the power prediction model matches the actual observed output power on the curve, the power prediction model is reliable; otherwise, it is unreliable.

[0082] Furthermore, the method for analyzing and processing the real-time climate data in step a23 includes:

[0083] D1: Extract satellite meteorological images over each forecast area (it should be noted that the air density ρ, which is the same as that in historical climate data, is also obtained here);

[0084] D2: Arrange the satellite meteorological images corresponding to each cycle in sequence to form an image dataset, and perform grayscale processing on the satellite meteorological images in the image dataset to identify the corresponding wind power equipment features and satellite cloud features. Based on the feature comparison method, obtain the changes in wind power equipment features and satellite cloud features, and calculate the changes in wind speed and wind direction to realize the analysis and processing of real-time climate data.

[0085] Furthermore, the feature comparison method in step D2 specifically includes:

[0086] d21: Overlay satellite meteorological images according to their chronological order, and compare the features of adjacent satellite meteorological images, including:

[0087] d22: Mark the center of the wind turbine's feature area as feature point K, and mark the center of the satellite cloud feature as feature point J. The selection of the satellite cloud feature should be consistent with the wind direction and the wind power receiving direction of the wind turbine. Only when the directions are consistent can the wind turbine effectively capture the wind and convert it into electrical energy.

[0088] d23: Overlay two adjacent satellite meteorological images. After overlay, the feature point K in the two adjacent satellite meteorological images will be in an overlapping state. Then, determine whether there is a deviation in wind direction or wind speed based on the position of feature point J.

[0089] Furthermore, the method for determining whether there is a deviation in wind direction or wind speed based on the position of feature point J in step d23 includes:

[0090] d231: Take the initial satellite meteorological image in the image dataset of one period as the base image. If there is a difference in the position of feature point J on the satellite meteorological image and the base image after overlay, mark the center of the wind power equipment feature area corresponding to the satellite meteorological image as feature point K', mark the center of the satellite cloud feature as feature point J', and determine whether feature point J' is located in the line segment direction between feature point K and feature point J.

[0091] d232: If feature point J' is located on the line segment direction between feature point K and feature point J, then the wind direction has not changed. Calculate the corresponding wind speed value within the period to determine whether the wind speed has changed.

[0092] d233: If feature point J' deviates from the line segment direction of feature point K and feature point J, the wind direction changes. Calculate the wind speed value along the line segment direction of feature point K and feature point J within the period to determine whether the wind speed has changed.

[0093] Furthermore, the methods for calculating the wind speed values ​​in steps d232 and d233, and the methods for determining whether the wind speed has changed, include:

[0094] The origin is set at the bottom left corner of the satellite weather map, and a coordinate system is formed with the bottom edge of the image as the X-axis and the left edge as the Y-axis.

[0095] The coordinates of feature point K and feature point J in the coordinate system are labeled as K(x1, y1) and J(x2, y2), and the coordinates of feature point K' and feature point J' are labeled as K'(x1', y1') and J'(x3, y3), where feature point K and feature point K' coincide.

[0096] The formula for calculating wind speed in step d232 is:

[0097]

[0098] Where V1 represents the wind speed value over one period. Let the length of the line segment between feature point K and feature point J be denoted as . It is represented by the line segment length between feature point K' and feature point J', m is the ratio of the distance on the image to the actual distance of the feature, and T is one period time;

[0099] The formula for calculating wind speed in step d233 is:

[0100]

[0101] Wherein, V2 represents the wind speed value within one period along the line segment between feature point K and feature point J after the wind direction changes. It represents the distance moved along the line segment direction between feature point K and feature point J, and α represents the wind direction deviation angle between the line segment direction between feature point K and feature point J.

[0102] V1 and V2 are compared with the initial wind speed. If the absolute value of the difference between V1 or V2 and the initial wind speed is within the deviation threshold, the wind speed has not changed. If the absolute value of the difference between V1 or V2 and the initial wind speed is outside the deviation threshold, the wind speed has changed, which will cause the output power of the wind turbine to change.

[0103] Furthermore, the method for introducing the analyzed and processed real-time climate data into the power prediction model in step a23 to obtain new prediction results includes:

[0104] E1: The wind speed data in the real-time climate data after analysis and processing is introduced into the power prediction model. The power prediction model generates a new predicted power and outputs it. If the new predicted power is greater than the rated output power of the wind power equipment, the rated output power of the wind power equipment is used as the new predicted power.

[0105] E2: The formula for calculating the new predicted power in step E1 is:

[0106]

[0107] Where P2 is the new predicted output power, C is the wind energy utilization coefficient of the wind turbine (a constant value), R is the radius of the wind turbine rotor sweep, ρ is the air density of the predicted area, and V new The wind speed value over one cycle (equivalent to V1 or V2);

[0108] The air density ρ changes with wind speed and is affected by the air pressure, temperature, and humidity of the predicted area. The calculation formula is as follows:

[0109]

[0110] Where F is the atmospheric pressure value of the predicted area, q is the temperature value of the predicted area, and f is the water vapor pressure value of the predicted area.

[0111] Applying the above prediction method to the actual power prediction operation in area A, the initial predicted wind speed is 5 m / s, the wind energy utilization factor of the wind turbine is 0.5, the air density ρ in the prediction area is 1.225 kg / m³ under standard atmospheric pressure, and the sweeping radius of the wind turbine rotor is 3 m. The calculated initial predicted power value is 1081.8 kW. However, after real-time data updates, the updated wind speed is 10 m / s, the wind energy utilization factor of the wind turbine is 0.5, the air density ρ in the prediction area is 1.225 kg / m³ under standard atmospheric pressure, and the sweeping radius of the wind turbine rotor is 3 m. The calculated predicted power value exceeds the rated output power of the wind turbine. Therefore, the predicted power is based on the rated output power of the wind turbine, which is 2500 kW.

[0112] This embodiment introduces real-time wind speed data into the power prediction model. The power prediction model generates and calculates new predicted power based on changes in wind speed and outputs it. This allows us to obtain information about the impact of wind speed on changes in predicted output power, as well as the changes in wind speed caused by changes in air density ρ. This leads to the prediction of changes in output power, ensuring that the predicted power result is within a predetermined deviation range of the actual output power. Furthermore, it makes the predicted result more closely match the actual power, thereby enabling practical application in real-world regions.

[0113] Furthermore, the method for comparing the new prediction result with the actual output power of the wind power equipment in step A3 includes:

[0114] A31: Extract the predicted power results obtained from multiple cycle power prediction models, and extract the actual output power of wind power equipment in the same cycle.

[0115] A32: The predicted power result is compared with the actual output power according to the cycle. If the predicted power result is within the predetermined deviation range of the actual output power, the predicted power result is accurate; otherwise, the predicted power result is inconsistent with the actual output power.

[0116] In this embodiment, taking a 10-minute cycle as an example, the predicted power data within 1000 cycles is compared with the output power of the wind power equipment within the corresponding 1000 cycles. If the predicted power value meets the output power ±2kW value of the wind power equipment, then the predicted power value is accurate. Based on the above comparison, 987 predicted power values ​​are accurate. Therefore, by using the new energy power prediction operation of this invention, the power prediction success rate within 1000 cycles reaches 98.7%, thereby achieving a relatively accurate new energy power prediction operation.

[0117] Example 2

[0118] This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the wind farm output power prediction method described in Embodiment 1.

[0119] Example 3

[0120] This embodiment provides a computer-readable storage medium storing a computer program thereon. When the computer program is executed by a processor, it implements the steps of the wind farm output power prediction method described in Embodiment 1.

[0121] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of this application can be implemented in various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0122] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0123] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0124] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0125] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0126] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for predicting wind farm output power, characterized in that: The following steps are involved: A1: Collect historical climate data, new energy power generation data and real-time climate data, wherein the new energy power generation data is the output power of wind power equipment; A2: extracting key features from the historical climate data, establishing a power prediction model based on the key features and the output power of the wind power equipment to predict the output power of the wind power equipment, and optimizing the power prediction model through the real-time climate data to achieve real-time update of the prediction result of the output power of the wind power equipment, specifically including: a21: preprocessing the historical climate data to extract key features; a22: Establish a power prediction model based on the extracted key features and the corresponding wind power equipment output power; a23: analyzing and processing the real-time climate data, and introducing the analyzed and processed real-time climate data into a power prediction model to obtain a new prediction result; A3: The accuracy of the prediction result is obtained by comparing the new prediction result with the actual output power of the wind power equipment.

2. The method for predicting wind farm output power according to claim 1, characterized in that: The method for preprocessing the historical climate data in step a21 includes: B1: Establish a matching template for required data, and set row and column headers for the matching template, wherein the row headers correspond to category names of different wind power equipment, and the column headers correspond to category names of different time periods or different scenarios; B2: Introduce the collected historical climate data into the matching template: first match the category name under the column title, then match the corresponding row title under the column title, extract the corresponding data content after the title matches the same, and record it in the data grid where the matched column title is located vertically and the row title is located horizontally; B3: After the matching template is matched, the data recorded by the matching template is retained to filter and clean up the unmatched data.

3. The method for predicting wind farm output power according to claim 1, characterized in that: The method for establishing a power prediction model according to the extracted key features and the corresponding wind power equipment output power in step a22 includes: C1: extracting the wind speed data in the key feature, and obtaining the output power data of n wind power equipment in the same area corresponding to the wind speed data; C2: Constructing a power prediction model based on the wind speed data and the corresponding wind power equipment output power data.

4. The method for predicting wind farm output power according to claim 3, characterized in that: The step C2 further includes: obtaining curve graph data on the wind speed and the average output power change of n wind power equipment based on the power prediction model, wherein the average output power calculation formula of n wind power equipment is: Where P1 is the predicted average output power of n wind power equipment in the same area, (p1+p1+...+p n ) is the total output power of n wind turbines in the same area per cycle, and n represents the number of wind turbines in the same area.

5. The method for predicting wind farm output power according to claim 2, characterized in that: The method for analyzing and processing the real-time climate data in step a23 includes: D1: Extract satellite weather images over each forecast area; D2: Arrange the satellite meteorological images corresponding to each period in order to form an image data set, and grayscale the satellite meteorological images in the image data set to identify the corresponding wind power equipment characteristics and satellite cloud layer characteristics. According to the feature comparison method, the changes in wind power equipment characteristics and satellite cloud layer characteristics are obtained, and the changes in wind speed and wind direction are calculated to realize the analysis and processing of real-time climate data.

6. The method for predicting wind farm output power according to claim 5, characterized in that: The feature comparison method in step D2 is specifically as follows: d21: Overlay satellite weather maps according to their time sequence and compare the features of two adjacent satellite weather maps, including: d22: Mark the center of the wind turbine feature area as feature point K, and mark the center of the satellite cloud feature as feature point J. The satellite cloud feature is selected to keep the wind direction consistent with the wind power receiving direction of the wind turbine; d23: Overlay two adjacent satellite weather maps. After the overlay, the feature point K in the two adjacent satellite weather maps is in a coincident state. Then, according to the position of the feature point J, it is determined whether the wind direction or wind speed has deviated.

7. The method for predicting wind farm output power according to claim 6, characterized in that: The method for determining whether the wind direction or wind speed has deviated according to the position of the feature point J in step d23 includes: d231: Take the initial satellite meteorological map in the image data set of one period as the basic image. If there is a difference in the position of the feature point J on the satellite meteorological map and the basic image after overlay, mark the center of the wind power equipment feature area corresponding to the satellite meteorological map as the feature point K', mark the center of the satellite cloud feature as the feature point J', and determine whether the feature point J' is located in the direction of the line segment between the feature point K and the feature point J; d232: If the feature point J' is located in the direction of the line segment between the feature point K and the feature point J, the wind direction has not changed at this time. The corresponding wind speed value within the cycle is calculated to determine whether the wind speed has changed; d233: If feature point J' deviates from the line segment direction between feature point K and feature point J, the wind direction changes. The wind speed value along the line segment direction between feature point K and feature point J within the period is calculated to determine whether the wind speed changes.

8. The method for predicting wind farm output power according to claim 7, characterized in that: The method for calculating the wind speed value in step d232 and step d233 and the method for determining whether the wind speed has changed include: Take the lower left corner of the satellite weather map as the origin, and form a coordinate system from the origin with the lower edge of the image as the X-axis and the left edge as the Y-axis; The coordinates of the feature point K and the feature point J in the coordinate system are marked as K(x1, y1) and J(x2, y2), and the coordinates of the feature point K' and the feature point J' are marked as K'(x1', y1') and J'(x3, y3), where the feature point K coincides with the feature point K'; The wind speed calculation formula in step d232 is: Among them, V1 represents the wind speed value within a cycle time. It is expressed as the length of the line segment between feature point K and feature point J. It is expressed as the length of the line segment between feature point K' and feature point J', m is the ratio of the distance on the image to the actual distance of the feature, and T is a cycle time; The wind speed calculation formula in step d233 is: Among them, V2 represents the wind speed value within one cycle along the line segment direction between feature points K and J after the wind direction changes. It is expressed as the moving distance along the line segment direction between feature point K and feature point J, and α represents the wind direction deviation angle between feature point K and feature point J; V1 and V2 are compared with the initial wind speed respectively. If the absolute value of the difference between V1 or V2 and the initial wind speed is within the deviation threshold, the wind speed has not changed; if the absolute value of the difference between V1 or V2 and the initial wind speed is outside the deviation threshold, the wind speed has changed, causing the output power of the wind power equipment to change.

9. The method for predicting wind farm output power according to claim 8, characterized in that: In step a23, the method of introducing the analyzed and processed real-time climate data into the power prediction model to obtain a new prediction result includes: E1: The wind speed data in the real-time climate data after analysis and processing is introduced into the power prediction model. The power prediction model generates and outputs a new predicted power. If the new predicted power is greater than the rated output power of the wind power equipment, the rated output power of the wind power equipment is used as the new predicted power. E2: The calculation formula for the new predicted power in step E1 is: Among them, P2 is the new predicted output power, C is the wind energy utilization coefficient of the wind turbine and is a constant value, R is the sweeping radius of the wind turbine rotor, ρ is the air density in the predicted area, V new is the wind speed value within a cycle time; The air density ρ is affected by the air pressure, temperature and humidity of the forecast area as the wind speed changes. The calculation formula is: Among them, F is the atmospheric pressure value of the prediction area, q is the temperature value of the prediction area, and f is the water vapor pressure value of the prediction area.

10. The method for predicting wind farm output power according to claim 1, characterized in that: The method for comparing the new prediction result with the actual output power of the wind power equipment in step A3 includes: A31: extract the predicted power results obtained by multiple period power prediction models, and extract the actual output power of the wind power equipment in the same corresponding period; A32: The predicted power result is compared with the actual output power in a periodic manner. When the predicted power result is within a predetermined deviation range of the actual output power, the predicted power result is accurate. Otherwise, the predicted power result is inconsistent with the actual output power.

11. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method for predicting output power of a wind farm as claimed in any one of claims 1 to 10 are implemented.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for predicting output power of a wind farm as claimed in any one of claims 1 to 10 are implemented.

Images