A meteorological optimization system and method based on an AI meteorological model

Abstract

The present invention discloses a meteorological optimization system and method based on an AI meteorological model, relating to the technical field of meteorological optimization, including: a data collection module, which includes a collection unit, a preprocessing unit and a data monitoring unit; a prediction module; the prediction module is used to obtain predicted weather data according to the meteorological data of the data collection module; a decision support module; if the data is defective or incomplete, affecting the accuracy of the simulation, so this technical solution monitors and judges whether the observed data is defective and provides high-quality observed data; it is also possible to determine the data sparse areas around the wind power equipment and then correspondingly improve the data coverage rate. According to the particularity within the area to be enhanced in monitoring, corresponding treatment plans can be formulated, taking into account not only the issue of improvement costs, but also enabling the data coverage rate of the area to be enhanced in monitoring to meet the requirements.

Description

Technical Field

[0001] The present invention relates to the technical field of meteorological optimization, and more specifically, to a meteorological optimization system and method based on an AI meteorological model. Background Art

[0002] With the construction of a power system mainly based on new energy becoming more and more perfect, due to the particularity of wind power equipment, it is generally installed in an environment with excellent wind resources. However, such environments are often accompanied by extreme disasters, which may pose challenges to the stable operation of wind power equipment.

[0003] Chinese Patent with application number 2022109068269 discloses a method and system for optimizing the layout of a power meteorological monitoring network. Among them, this method conducts a data assimilation numerical simulation experiment of simulation monitoring data through control simulation experiments and various sensitivity simulation experiments to determine the results of the data assimilation numerical simulation experiment of simulation monitoring data; according to the results of the data assimilation numerical simulation experiment of simulation monitoring data, combined with real meteorological field data, it conducts an evaluation and verification of the data assimilation effect to determine the optimal layout plan of the power meteorological monitoring network.

[0004] However, since the existing meteorological monitoring network is not yet mature and mainly relies on the meteorological monitoring of the meteorological department, although there are ways to optimize the meteorological layout through simulation experiments and data, data assimilation depends on high-quality, high spatio-temporal resolution observation data. If the data is defective or incomplete, it may affect the accuracy of the simulation; the model may perform well in a specific power grid area, but its generalization ability is insufficient in other areas or different climate conditions. In rapidly changing meteorological conditions, there will also be a phenomenon of untimely response of the monitoring network layout to real-time data, affecting the accuracy of the judgment results. Summary of the Invention

[0005] To solve the above problems, the present invention provides a meteorological optimization system and method based on an AI meteorological model.

[0006] To achieve the above purpose, on the one hand, the present invention adopts the following technical solutions:

[0007] A meteorological optimization system based on an AI meteorological model, comprising:

[0008] A data collection module, which includes a collection unit, a preprocessing unit, and a data monitoring unit;

[0009] The collection unit is used to collect meteorological data of the environment around the wind power equipment;

[0010] The preprocessing unit is used to preprocess the collected meteorological data; the data monitoring unit is used to judge whether there are defects in the meteorological data obtained by the collection unit, specifically:

[0011] Obtain the statistical characteristic defect value of the meteorological data and mark it as Q;

[0012] Obtain the time consistency defect value of the meteorological data and mark it as W;

[0013] Obtain the spatial consistency defect value of the meteorological data and mark it as E;

[0014] According to the formula , calculate and obtain the defect value of the meteorological data, where , and are preset weight coefficients;

[0015] Preset a threshold for the defect value R in advance, and compare the defect value R with the threshold. If the defect value R is greater than the threshold, it is determined that the meteorological data has defects. If the defect value R is less than or equal to the threshold, it is determined that the meteorological data is complete;

[0016] If the meteorological data has defects, re-collect it until the meteorological data is complete;

[0017] Prediction module; the prediction module is used to obtain the predicted weather data according to the meteorological data of the data collection module;

[0018] Decision support module; the decision support module is used to assist the user in understanding the prediction result according to the weather data within a future period of time obtained by the prediction module.

[0019] Preferably, the acquisition unit is further used to enhance the data coverage rate around the wind power equipment, specifically:

[0020] Taking the position of the wind power equipment as the origin, draw a circle with a preset radius to obtain the monitoring range of the meteorological data;

[0021] Divide the monitoring range into Y regions on average according to the area; obtain the number of observation stations in each region and mark it as K, mark the center point of each region as the point to be estimated, and mark the other observation stations in the region as adjacent points; according to the formula Calculate and obtain the meteorological parameters in each region, where is the meteorological parameter of the th adjacent point, d is the distance between the point to be estimated and the th adjacent point, is the power of the distance weight;

[0022] Set a threshold for the meteorological parameter S within a preset area, marked as the first threshold, and calculate the difference between the meteorological parameter S of the point to be estimated in each area and the first threshold;

[0023] Set a threshold for the difference, marked as the second threshold, and compare the difference between the meteorological parameter S of the point to be estimated in each area and the threshold with the second threshold. Mark the areas where the difference between the meteorological parameter S of the point to be estimated in all areas and the threshold is greater than the second threshold as the areas to be enhanced in monitoring;

[0024] Perform corresponding processing on the areas to be enhanced in monitoring to improve data coverage.

[0025] Preferably, the specific manner in which the acquisition unit performs corresponding processing on the areas to be enhanced in monitoring is as follows:

[0026] According to the formula Obtain the coverage gap score within the area to be enhanced in monitoring ; where is the preset target coverage rate, is the actual coverage rate within the area to be enhanced in monitoring;

[0027] According to the formula Obtain the sensitivity score of the observation stations within the area to be enhanced in monitoring , where is the maximum improvement percentage;

[0028] According to the formula , calculate and obtain the comprehensive data sparsity index within the area to be enhanced in monitoring , where and are preset proportionality coefficients;

[0029] Set a threshold for the data sparsity index and judge whether it is less than the threshold. If so, move other observation stations within the area. If not, set up new observation stations.

[0030] Preferably, the manner in which the acquisition unit moves other observation stations within the area to be enhanced in monitoring is as follows. Specifically:

[0031] Obtain the movement priority value of the observation stations within the area to be enhanced in monitoring ;

[0032] Sort the observation stations within the area to be enhanced in monitoring in descending order according to the movement priority value , and preferentially select the observation station with the largest movement priority value ;

[0033] For the movement priority value The largest observation station adjusts its position, and then calculates the difference between the meteorological parameter S of the point to be estimated in the monitoring area to be enhanced after adjustment and the threshold value. The difference between the meteorological parameter S and the threshold value is compared with the threshold value of the difference. If the difference between the meteorological parameter S and the threshold value is less than the threshold value of the difference, the adjustment is stopped; otherwise, the adjustment continues until the difference between the meteorological parameter S and the threshold value is less than the threshold value of the difference.

[0034] Preferably, according to the formula the mobile priority value is calculated and obtained , where is the number of observation stations in the monitoring area to be enhanced, is the distance between the c-th observation device and the point to be estimated, is the observation radius of the c-th observation device, is a preset exponential function, which is 1 when the condition is satisfied and 0 when it is not satisfied.

[0035] Preferably, the acquisition unit increases the observation stations in the monitoring area to be enhanced in the following manner, specifically:

[0036] According to the formula the number of additional stations t in the monitoring area to be enhanced is calculated and obtained, where is the coverage area of the monitoring station obtained by taking the monitoring station as the center and the preset monitoring distance as the radius.

[0037] Preferably, the prediction module is used to obtain the predicted weather data according to the meteorological data of the data collection module, specifically:

[0038] The adjusted meteorological parameter is brought into the training model to generate the weather information for a future period of time.

[0039] Preferably, the decision support module is used to assist the user in understanding the prediction result according to the weather data for a future period of time obtained by the prediction module, specifically:

[0040] The decision support module includes a user interaction unit, and the user interaction unit is used to obtain the prediction result of the prediction module and transmit and display the prediction result and the historical data.

[0041] On the other hand, the present invention also proposes a meteorological optimization method based on an AI meteorological model, including the following steps:

[0042] Step 1: Collect the meteorological data of the environment around the wind power equipment and determine whether there are defects in the meteorological data;

[0043] Step 2: Obtaining predicted weather data based on the meteorological data of the data collection module;

[0044] Step 3: Obtain weather data for a period of time in the future based on the prediction module to assist users in understanding the prediction results.

[0045] Beneficial effects: By setting up a data collection module, if the data is defective or incomplete, it will affect the accuracy of the simulation. Therefore, this technical solution determines whether the observation data has defects through monitoring and provides high-quality observation data; it can also determine the data-sparse area around the wind power equipment, and then improve the data coverage accordingly, add observation sites in the monitoring area to be enhanced with insufficient data, and move other observation sites in the area. According to the particularity of the monitoring area to be enhanced, a corresponding processing plan can be formulated, which not only takes into account the cost of improvement, but also ensures that the data coverage of the monitoring area to be enhanced meets the requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a system block diagram of the present invention.

[0047] Figure 2 The present invention is a flow chart of the method. DETAILED DESCRIPTION

[0048] It should be noted that the meteorological optimization system is used to provide accurate weather forecasts during the use of wind power equipment, so that wind farms can arrange the operation of wind power equipment more effectively and improve power generation efficiency. It can also help predict extreme weather times, such as thunderstorms, strong winds and freezing, which is conducive to wind power equipment to take preventive measures in advance and reduce equipment damage. However, in actual use, data assimilation may rely on high-quality, high-temporal and high-resolution observation data, because data assimilation refers to combining observation data with numerical models to generate the best estimate of the environmental state. This process includes not only simple extrapolation of observation data, but also involves interpretation of data and adjustment of models to ensure that the final estimate is as close to the actual state as possible. If the data is defective or incomplete, it affects the accuracy of the simulation. Therefore, this technical solution monitors and determines whether the observation data is defective and provides high-quality observation data to solve the above problems, as follows:

[0049] like Figure 1 Shown: A meteorological optimization system based on an AI meteorological model, comprising:

[0050] A data collection module, the data collection module includes a collection unit, a preprocessing unit and a data monitoring unit;

[0051] The acquisition unit is used to collect meteorological data of the environment around the wind power equipment; it should be noted that in this embodiment, the sources of meteorological data can include ground observation data, satellite remote sensing data, radar data, and upper-air observation data, and the meteorological data specifically includes air temperature, dew point temperature, air pressure, wind speed, wind direction, precipitation, cloud cover, visibility, etc.;

[0052] It should also be noted that ground observation data are all obtained through ground equipment monitoring, and there will inevitably be areas with sparse data monitoring in the face of complex terrain;

[0053] The preprocessing unit is used to preprocess the collected meteorological data; it should be noted that in this embodiment, the preprocessing method can be data cleaning, and the purpose is to remove incorrect or inconsistent entries in the data;

[0054] The data monitoring unit is used to judge whether there are defects in the meteorological data obtained by the acquisition unit, specifically:

[0055] Obtain the statistical characteristic defect value of the meteorological data and mark it as Q; it should be noted that the statistical characteristic defect value Q is used to measure the fluctuation degree of the meteorological data. In this embodiment, through the formula: Calculate and obtain, where is the standard deviation of the daily meteorological data within the current month, is the mean value of the meteorological data within the current month;

[0056] Obtain the time consistency defect value of the meteorological data and mark it as W; it should be noted that the time consistency defect value W can be measured by the deviation of the autocorrelation coefficient. In this embodiment, the time consistency defect value W is calculated and obtained through the difference between the autocorrelation coefficient and the expected correlation coefficient, specifically:

[0057]

[0058] In this embodiment, the actual autocorrelation coefficient is a measure of the mutual relationship of time series data at different time lags. The acquisition method of the autocorrelation coefficient is , where is the meteorological data value at time point t, is the mean value of the meteorological data within the current month, is the total number of observations in the current month, is the lag time, and the lag time is set in advance by the staff and the unit is hours;

[0059] The expected correlation coefficient can be obtained by the staff based on historical data. Specifically, in this embodiment, it can be:

[0060] By analyzing historical data at the same monitoring site or under similar climatic conditions, calculate the historical autocorrelation coefficient as the expectation Alternatively, a long-term average autocorrelation coefficient can be calculated based on multi-year observation data as the expectation ;

[0061] The time consistency defect value W is used to measure the continuity and stability of meteorological data in the time series. By analyzing the change of the autocorrelation coefficient, it can assist the staff to quickly discover possible missing or discontinuous areas in the meteorological data;

[0062] It should also be noted that by calculating the time consistency defect value W, the abnormal fluctuation degree of the same site in different time periods can be compared to assist in judging the accuracy of the data;

[0063] Obtain the spatial consistency defect value of the meteorological data and mark it as E; it should be noted that

[0064] It should be noted that the spatial consistency defect value E is used to measure the continuity and stability of meteorological data in the time series, reflecting whether the data points conform to the expected meteorological processes and patterns in time. In this embodiment, the acquisition method of the spatial consistency defect value E can be: According to the formula:

[0065] Obtain the spatial consistency defect value E of the meteorological data, where is the numerical value of the number of the meteorological data observation areas, is the actual observed value, is the estimated value obtained by the staff through the spatial interpolation method;

[0066] According to the formula Calculate and obtain the defect value of the meteorological data ;

[0067] Where , and are preset weight coefficients; it should be noted that in this embodiment , and are formulated according to the actual situation and importance. In this embodiment, , and can take the values of 0.128, 0.398 and 0.474;

[0068] It should also be noted that according to the experimental data, there is a close proportional relationship between the defect value of meteorological data, the defect value of statistical characteristics, the defect value of time consistency, and the defect value of spatial consistency. The defect value of meteorological data can be obtained through comprehensive calculation. , and because meteorological data may depend on high-quality, high spatio-temporal resolution observation data for data assimilation, if the data is defective or incomplete, it will affect the accuracy of simulation. Therefore, by obtaining the defect value of meteorological data, it can be judged whether there are defects in the meteorological data;

[0069] A threshold value of the defect value R is set in advance, and the defect value R is compared with the threshold value. If the defect value R is greater than the threshold value, it is judged that the meteorological data is defective. If the defect value R is less than or equal to the threshold value, it is judged that the meteorological data is complete; it should be noted that by detecting the defect value R of the meteorological data, it can be judged whether there are defects in the observation data, thereby improving the accuracy of subsequent prediction results;

[0070] If the meteorological data is defective, it will be recollected until the meteorological data is complete;

[0071] Prediction module; the prediction module is used to obtain the predicted weather data according to the meteorological data of the data collection module;

[0072] Decision support module; the decision support module is used to assist users in understanding the prediction results according to the weather data in a future period of time obtained by the prediction module.

[0073] As an optional embodiment, it should be noted that although the existing data collection methods can obtain meteorological information from multiple data sources, since wind power equipment is installed in relatively remote areas in order to utilize wind power resources, there will be areas with sparse data monitoring. This technical solution can solve the above problems by determining the data sparse areas around the wind power equipment and then correspondingly improving the data coverage rate;

[0074] The acquisition unit is also used to enhance the data coverage rate around the wind power equipment, specifically:

[0075] Taking the position of the wind power equipment as the origin and drawing a circle with a preset radius to obtain the monitoring range of the meteorological data;

[0076] The monitoring range is evenly divided into Y areas according to the area; it should be noted that in this embodiment, the value of Y is evaluated by the staff according to the area size of the monitoring range;

[0077] Obtain the number of observation stations in each area, denoted as K, mark the center point in each area as the point to be estimated, and mark the other observation stations in the area as adjacent points; it should be noted that in this embodiment, the observation stations can be fixed meteorological monitoring stations and mobile meteorological monitoring vehicles for obtaining meteorological data;

[0078] According to the formula Calculate and obtain the meteorological parameters in each area , where is the meteorological parameter of the th adjacent point, d is the distance between the point to be estimated and the th adjacent point, is the power of the distance weight;

[0079] It should be noted that the meteorological parameters of the adjacent points are obtained by the staff in advance. The specific calculation method can be to collect the daily average temperature, precipitation, wind speed, relative humidity and atmospheric pressure of the adjacent point for calculation. The further calculation method can be:

[0080]

[0081] where , , , and are preset proportionality coefficients, which are obtained according to historical data in this embodiment, and the values can be , , , , and ;

[0082] It should be noted that in this embodiment, the value of the power is 2, which is a commonly used default value; it should be noted that the meteorological parameters in an area can be calculated through the meteorological parameters of the point to be estimated and the surrounding adjacent points. In the case of uneven distribution of observation stations, the appropriate power and the adjacent points to be estimated in the driving area can also be selected to flexibly obtain the meteorological parameters of the point to be estimated in each area;

[0083] Set a threshold for the meteorological parameter S in an area in advance, denoted as the first threshold, and calculate the difference between the meteorological parameter S of the point to be estimated in each area and the first threshold;

[0084] A threshold value of the difference is preset, marked as the second threshold value, and the difference between the meteorological parameter S of the point to be estimated in each region and the threshold value is compared with the second threshold value. The regions where the difference between the meteorological parameter S of the point to be estimated in all regions and the threshold value is greater than the second threshold value are marked as regions to be enhanced for monitoring; it should be noted that through calculation, it can be determined that the regions to be enhanced for monitoring are also data-sparse regions, which can assist in improving data coverage and quality;

[0085] Corresponding processing is performed on the regions to be enhanced for monitoring to improve data coverage.

[0086] As an optional embodiment, it should be noted that after determining the regions to be enhanced for monitoring, a method for improving the coverage rate of the regions to be enhanced for monitoring needs to be obtained. Adding observation stations in the regions to be enhanced for monitoring with insufficient data is the most direct enhancement means. However, it is very difficult to determine the number and location of the added observation stations, and because the cost of adding stations is relatively high, other observation stations in the region can also be moved. This technical solution can formulate corresponding processing plans according to the particularity within the regions to be enhanced for monitoring, not only considering the issue of improvement cost, but also enabling the data coverage rate of the regions to be enhanced for monitoring to meet the requirements. Specifically:

[0087] The specific manner in which the acquisition unit performs corresponding processing on the regions to be enhanced for monitoring is as follows:

[0088] According to the formula The coverage rate gap score within the regions to be enhanced for monitoring is obtained ; where is the preset target coverage rate, is the actual coverage rate within the regions to be enhanced for monitoring; it should be noted that the preset target coverage rate is obtained based on the regional area combined with historical data, and the actual coverage rate is obtained by dividing the total area covered by the monitoring stations by the total area of the regions to be enhanced for monitoring. The acquisition method of the total area covered by the monitoring stations can be to obtain the coverage area of the monitoring stations with the monitoring station as the center and the preset monitoring distance as the radius. Therefore, the combined coverage area of the monitoring stations is the total area covered by the monitoring stations;

[0089] According to the formula The sensitivity score of the observation stations within the regions to be enhanced for monitoring is obtained where is the maximum improvement percentage; it should be noted that in this embodiment, the maximum improvement percentage is obtained through experiments. Specifically: First, determine the indicators used to measure data quality, such as mean square error (MSE), coefficient of determination, mean absolute error, etc.; then create different simulation scenarios, with one or more observation stations added in each scenario, and these scenarios can be achieved by changing the positions of existing stations or adding new stations; after that, in each simulation scenario, use existing interpolation methods or other statistical methods to simulate the meteorological parameters of the entire monitoring area; for each simulation scenario, calculate the change in the data quality indicator, that is, the difference between the indicator value in the simulation scenario and the indicator value in the baseline scenario (without adding observation stations); analyze the patterns of indicator changes in different simulation scenarios to determine which areas adding stations can improve data quality; the maximum improvement percentage refers to the percentage value with the largest improvement in the data quality indicator among all simulation scenarios. In this embodiment, the value of the largest percentage is 0.12;

[0090] According to the formula , calculate and obtain the comprehensive data sparsity index in the to-be-enhanced monitoring area , where and are preset proportionality coefficients; it should be noted that in this embodiment, and can take the values of 0.476 and 0.514; it also should be noted that according to experiments, the data sparsity index has a proportional relationship with the coverage gap score and the sensitivity score. By calculating the data sparsity index , the data sparsity degree in the to-be-enhanced monitoring area can be intuitively judged. In this embodiment, if N is close to 1, it means that the data coverage quality in the to-be-enhanced monitoring area is good. If N is much lower than 1, it means that the data is sparse;

[0091] Preset a threshold for the data sparsity index, and judge whether is less than the threshold. If so, move other observation stations in the area. If not, set up new observation stations. It should be noted that by using the data sparsity index of the to-be-enhanced monitoring area in the moving area to judge the processing method for the to-be-enhanced monitoring area, for the to-be-enhanced monitoring area with relatively sparse data, the method of setting up new stations with better effects is adopted. For the to-be-enhanced monitoring area with slightly sparse data, the method of moving other observation stations in the area to save costs is adopted, which can not only meet the usage requirements but also minimize costs as much as possible;

[0092] As an optional embodiment, it should be noted that after determining the data sparsity of the monitoring area to be enhanced in the moving area and making a processing method for other observation sites in the moving area, it is difficult to determine the observation site that needs to be moved in a short time. This technical solution can quickly determine the observation site that needs to be moved by enhancing the environment in the monitoring area, which can not only increase the accuracy of the adjustment, but also reduce the prediction time;

[0093] The manner in which the acquisition unit moves other observation sites within the to-be-enhanced monitoring area is as follows:

[0094] Obtain the mobility priority value of the observation site in the enhanced monitoring area ;

[0095] The observation stations in the enhanced monitoring area are assigned according to the mobile priority value. Sort by size, giving priority to mobile priority values The largest observation site;

[0096] Mobile priority value The largest observation station adjusts its position, and then calculates the difference between the meteorological parameter S and the threshold of the point to be estimated in the enhanced monitoring area after adjustment, and compares the difference between the meteorological parameter S and the threshold with the threshold of the difference. If the difference between the meteorological parameter S and the threshold is less than the threshold of the difference, the adjustment is stopped; if not, the adjustment is continued until the difference between the meteorological parameter S and the threshold is less than the threshold of the difference.

[0097] As an optional embodiment, the mobile priority value The method of obtaining is as follows:

[0098] According to the formula Calculate and obtain the mobile priority value ,in is the number of observation stations in the monitoring area to be enhanced, is the distance between the cth observation device and the point to be estimated, is the observation radius of the cth observation device, is a preset exponential function, which is 1 when the condition is met and 0 when it is not met. It can be obtained by combining the data sparse index and the radius of the observation device. In actual use, the mobile priority value The largest observation site is the one with the most overlapped monitoring radius. Prioritizing adjustment can make full use of the equipment and save costs.

[0099] As an optional embodiment, the collection unit increases the observation sites in the to-be-enhanced monitoring area in the following manner, specifically:

[0100] According to the formula calculate and obtain the number of additional stations in the to-be-enhanced monitoring area, where is the coverage area of the monitoring station obtained by taking the monitoring station as the center of the circle and the preset monitoring distance as the radius.

[0101] As an optional embodiment, the prediction module is used to obtain predicted weather data according to the meteorological data of the data collection module, specifically:

[0102] Bring the adjusted meteorological parameters into the training model to generate weather information for a period of time in the future; it should be noted that, in this embodiment, the training model can be the GFS model or the ECMWF model.

[0103] As an optional embodiment, the decision support module is used to assist the user in understanding the prediction result according to the weather data obtained by the prediction module for a period of time in the future, specifically:

[0104] The decision support module includes a user interaction unit, and the user interaction unit is used to obtain the prediction result of the prediction module and transmit and display the prediction result and historical data.

[0105] On the other hand, the present invention also proposes a meteorological optimization method based on an AI meteorological model, including the following steps:

[0106] Step 1: Collect meteorological data of the environment around the wind power equipment and determine whether there are defects in the meteorological data;

[0107] Step 2: Obtain predicted weather data according to the meteorological data of the data collection module;

[0108] Step 3: Assist the user in understanding the prediction result according to the weather data obtained by the prediction module for a period of time in the future.

[0109] Working principle

[0110] By setting a data collection module, if the data is defective or incomplete, it will affect the accuracy of the simulation. Therefore, this technical solution monitors and judges whether there are defects in the observed data and provides high-quality observed data; it is also possible to determine the data sparse area around the wind power equipment, and then correspondingly improve the data coverage rate, add observation stations in the to-be-enhanced monitoring area with insufficient data, or move other observation stations within the area. According to the particularity within the to-be-enhanced monitoring area, corresponding treatment plans can be formulated, which not only takes into account the issue of improvement cost, but also enables the data coverage rate in the to-be-enhanced monitoring area to meet the requirements.

[0111] The above are only the preferred embodiments of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions falling within the concept of the present invention belong to the protection scope of the present invention. It should be noted that for those of ordinary skill in the art, several improvements and refinements made without departing from the principle of the present invention should also be regarded as within the protection scope of this template.

Claims

1. A meteorological optimization system based on an AI meteorological model, characterized in that, Including: A data collection module, which includes a collection unit, a preprocessing unit, and a data monitoring unit; The collection unit is used to collect meteorological data of the environment around the wind power equipment; The preprocessing unit is used to preprocess the collected meteorological data; the data monitoring unit is used to judge whether there are defects in the meteorological data obtained by the collection unit. Specifically: Obtain the statistical characteristic defect value of the meteorological data and mark it as Q; Obtain the time consistency defect value of the meteorological data and mark it as W; Obtain the spatial consistency defect value of the meteorological data and mark it as E; According to the formula , calculate and obtain the defect value of the meteorological data , where , and are preset weight coefficients; Set a threshold value of a defect value R in advance, and compare the defect value R with the threshold value. If the defect value R is greater than the threshold value, it is judged that the meteorological data has defects. If the defect value R is less than or equal to the threshold value, it is judged that the meteorological data is complete; If the meteorological data has defects, re-collect it until the meteorological data is complete; A prediction module; the prediction module is used to obtain predicted weather data according to the meteorological data of the data collection module; A decision support module; The decision support module is used to assist users in understanding the prediction results according to the weather data in a future period obtained by the prediction module; The collection unit is also used to enhance the data coverage rate around the wind power equipment. Specifically: Taking the position of the wind power equipment as the origin, draw a circle with a preset radius to obtain the monitoring range of the meteorological data; Evenly divide the monitoring range into Y regions according to the area; Obtain the number of observation stations in each area, denoted as K, mark the center point in each area as the point to be estimated, and mark the other observation stations in the area as adjacent points; according to the formula Calculate and obtain the meteorological parameters in each area , where is the meteorological parameter of the th adjacent point, d is the distance between the point to be estimated and the th adjacent point, is the power of the distance weight; Set a threshold value of the meteorological parameter S in a region in advance, marked as the first threshold value, and calculate the difference between the meteorological parameter S of the point to be estimated in each region and the first threshold value; Set a threshold value of the difference in advance, marked as the second threshold value, and compare the difference between the meteorological parameter S of the point to be estimated in each region and the threshold value with the second threshold value. Mark the regions where the difference between the meteorological parameter S of the point to be estimated in all regions and the threshold value is greater than the second threshold value as the regions to be enhanced in monitoring; Perform corresponding processing on the regions to be enhanced in monitoring to improve the data coverage rate; The specific method for the collection unit to perform corresponding processing on the regions to be enhanced in monitoring is as follows: According to the formula obtain the coverage gap score within the to-be-enhanced monitoring area ; where is the pre-set target coverage rate, is the actual coverage rate within the to-be-enhanced monitoring area; According to the formula obtain the sensitivity scores of the observation stations in the monitoring area to be enhanced , where is the maximum improvement percentage; According to the formula , calculate and obtain the comprehensive data sparsity index within the to-be-enhanced monitoring area, where and are preset proportionality coefficients; Set a threshold for the data sparsity index in advance and judge whether it is less than the threshold. If so, move other observation stations within the area; if not, set up new observation stations.

2. The meteorological optimization system based on the AI meteorological model according to claim 1, characterized in that, The specific method for the collection unit to move other observation stations in the regions to be enhanced in monitoring is as follows: Obtain the mobile priority value of the observation stations within the to-be-enhanced monitoring area ; Sort the observation stations in the to-be-enhanced monitoring area according to the movement priority value in descending order, and preferentially select the observation station with the largest movement priority value; Adjust the position of the largest observation station for the mobile priority value Then calculate the difference between the meteorological parameter S of the point to be estimated in the monitoring area to be enhanced after adjustment and the threshold value. Compare the difference between the meteorological parameter S and the threshold value with the threshold value of the difference. If the difference between the meteorological parameter S and the threshold value is less than the threshold value of the difference, stop the adjustment; otherwise, continue the adjustment until the difference between the meteorological parameter S and the threshold value is less than the threshold value of the difference.

3. The meteorological optimization system based on the AI meteorological model according to claim 2, characterized in that, According to the formula calculate to obtain the mobile priority value , where is the number of observation stations in the monitoring area to be enhanced, is the distance between the c-th observation device and the point to be estimated, is the observation radius of the c-th observation device, is a preset exponential function, which is 1 when the condition is satisfied and 0 when the condition is not satisfied.

4. An optimized meteorological system based on an AI meteorological model according to claim 3, characterized in that, The specific method for the collection unit to increase the observation stations in the regions to be enhanced in monitoring is as follows: According to the formula calculate and obtain the number t of additional stations in the to-be-enhanced monitoring area, where is the coverage area of the monitoring station obtained by taking the monitoring station as the center and the preset monitoring distance as the radius.

5. The meteorological optimization system based on the AI meteorological model according to claim 1, characterized in that The prediction module is used to obtain predicted weather data according to the meteorological data of the data collection module. Specifically: Bring the adjusted meteorological parameters into the training model to generate weather information for a period of time in the future.

6. The meteorological optimization system based on the AI meteorological model according to claim 5, characterized in that, The decision support module is used to assist users in understanding the prediction results according to the weather data in a future period obtained by the prediction module. Specifically: The decision support module includes a user interaction unit, and the user interaction unit is used to obtain the prediction results of the prediction module and transmit and display the prediction results and historical data.

7. A meteorological optimization method based on an AI meteorological model, characterized in that, Including the following steps: Step 1: Collect meteorological data of the environment around the wind power equipment and judge whether there are defects in the meteorological data; Step 2: Obtain predicted weather data according to the meteorological data of the data collection module; Step 3: Assist the user in understanding the prediction result based on the weather data obtained within a certain period in the future by the prediction module; The specific process of Step 1 is as follows: Taking the location of the wind power equipment as the origin, draw a circle with a preset radius to obtain the monitoring range of the meteorological data; Evenly divide the monitoring range into Y regions according to the area; Obtain the number of observation stations in each area, denoted as K, mark the center point in each area as the point to be estimated, and mark the other observation stations in the area as adjacent points; According to the formula Calculate and obtain the meteorological parameters in each area , where is the meteorological parameter of the th adjacent point, d is the distance between the point to be estimated and the th adjacent point, is the power of the distance weight; Set a threshold of the meteorological parameter S in a region in advance, marked as the first threshold, and calculate the difference between the meteorological parameter S of the point to be estimated in each region and the first threshold; Set a threshold of the difference in advance, marked as the second threshold, and compare the difference between the meteorological parameter S of the point to be estimated in each region and the threshold with the second threshold. Mark the regions where the difference between the meteorological parameter S of the point to be estimated in all regions and the threshold is greater than the second threshold as the regions to be enhanced for monitoring.

Images