Solar radiation intensity short-term prediction method, device, equipment and storage medium

Abstract

The invention discloses a solar radiation intensity short-term prediction method, device and equipment and a storage medium. The method comprises the following steps: in historical meteorological element reanalysis data and historical satellite observation data, matching target reanalysis data and target satellite observation data corresponding to a ground station based on position information of the ground station; training the short-term radiation prediction model based on the target reanalysis data, the target satellite observation data and the historical real solar radiation data to obtain a trained short-term radiation prediction model; and based on the current surface meteorological element data, the current satellite observation data and the trained short-term radiation prediction model, predicting the solar radiation intensity of the ground station in a future short term to obtain a prediction result. According to the technical scheme, the prediction precision and accuracy of the solar radiation intensity in a short period can be improved, so that the corresponding solar resources can be accurately predicted, a solar power station is better helped to perform reasonable scheduling, and the power generation efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy power generation, and particularly to a short-term prediction method, device, equipment and storage medium for solar irradiance. Background Art

[0002] As a clean energy source, solar energy has been widely used in many fields such as power generation, heating and lighting. Accurate short-term and very short-term prediction of solar radiation is crucial for the balanced scheduling of electricity loads and the operation of power grids.

[0003] In the prior art, it is highly feasible and simple to calculate to estimate the solar radiation of a single station by using traditional methods based on experience such as sunshine hours and clear sky index. However, such methods often ignore other influencing factors such as humidity and temperature. The method itself has limited capacity for accommodating input data, and when the single-station model is extended to other stations or even the entire region for calculation, due to the differences in geographical and meteorological characteristics at each point within the region, the error will increase, and the spatial coverage ability of the model is limited. Although numerical model forecasts can provide data with global coverage, their spatio-temporal resolution is relatively coarse and the accuracy is low. Therefore, there is an urgent need for a short-term prediction method for solar radiation intensity to solve the above technical problems. Summary of the Invention

[0004] In view of this, the present invention provides a short-term prediction method, device, equipment and storage medium for solar radiation intensity, which can improve the prediction accuracy and precision of solar radiation intensity in the short term, so as to accurately predict the corresponding solar energy resources, better assist the reasonable scheduling of solar power plants, and improve the power generation efficiency.

[0005] According to one aspect of the present invention, an embodiment of the present invention provides a short-term prediction method for solar radiation intensity, the method comprising:

[0006] Obtain the historical true solar radiation data corresponding to each ground station in the target area during the historical preset time period, and obtain the historical meteorological element reanalysis data and historical satellite observation data during the historical preset time period from at least two third-party databases;

[0007] In the historical meteorological element reanalysis data and the historical satellite observation data, respectively match the target reanalysis data and the target satellite observation data corresponding to the ground station based on the position information of the ground station;

[0008] Train a short-term radiation prediction model based on the target reanalysis data, the target satellite observation data and the historical true solar radiation data to obtain a trained short-term radiation prediction model; wherein, the short-term radiation prediction model is composed of an XGBoost model;

[0009] Obtain the current surface meteorological element data and the current satellite observation data, and predict the solar radiation intensity of the ground station in the short term in the future based on the current surface meteorological element data, the current satellite observation data, and the trained short-term radiation prediction model to obtain a prediction result.

[0010] According to another aspect of the present invention, an embodiment of the present invention further provides a short-term prediction device for solar radiation intensity. The device includes:

[0011] A data acquisition module, configured to acquire the historical true solar radiation data corresponding to each ground station in the target area during a historical preset time period, and acquire the historical meteorological element reanalysis data and the historical satellite observation data during the historical preset time period from at least two third-party databases;

[0012] A matching module, configured to respectively match the target reanalysis data and the target satellite observation data corresponding to the ground station based on the position information of the ground station in the historical meteorological element reanalysis data and the historical satellite observation data;

[0013] A model training module, configured to train a short-term radiation prediction model based on the target reanalysis data, the target satellite observation data, and the historical true solar radiation data to obtain a trained short-term radiation prediction model; wherein, the short-term radiation prediction model is composed of an XGBoost model;

[0014] A prediction module, configured to obtain the current surface meteorological element data and the current satellite observation data, and predict the solar radiation intensity of the ground station in the short term in the future based on the current surface meteorological element data, the current satellite observation data, and the trained short-term radiation prediction model to obtain a prediction result.

[0015] According to another aspect of the present invention, an embodiment of the present invention further provides an electronic device. The electronic device includes:

[0016] At least one processor; and

[0017] A memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores a computer program executable by the at least one processor. The computer program is executed by the at least one processor so that the at least one processor can execute the short-term prediction method for solar radiation intensity according to any embodiment of the present invention.

[0019] According to another aspect of the present invention, an embodiment of the present invention further provides a computer-readable storage medium storing computer instructions for causing a processor to implement the short-term solar radiation intensity prediction method according to any embodiment of the present invention when executed.

[0020] According to another aspect of the present invention, an embodiment of the present invention further provides a computer program product, characterized in that the computer program product includes a computer program that implements the short-term solar radiation intensity prediction method according to any embodiment of the present invention when executed by a processor.

[0021] The technical effect of the present invention is that by fusing historical meteorological element reanalysis data and historical satellite observation data, in the historical meteorological element reanalysis data and historical satellite observation data, the target reanalysis data and target satellite observation data corresponding to the ground station are respectively matched based on the position information of the ground station. Then, the short-term radiation prediction model is trained by the target reanalysis data, target satellite observation data, and historical true solar radiation data to obtain a trained short-term radiation prediction model. On this basis, the solar radiation intensity of the ground station in the short-term future is predicted based on the current surface meteorological element data, current satellite observation data, and the trained short-term radiation prediction model to obtain a prediction result, which can improve the prediction accuracy and accuracy of the solar radiation intensity in the short term. Especially in the case of cloud cover and sudden weather changes, it can provide more accurate forecast results, so as to accurately predict the corresponding solar energy resources, better help solar power plants for reasonable scheduling, and improve power generation efficiency.

[0022] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0024] Figure 1 It is a flowchart of a short-term solar radiation intensity prediction method provided by an embodiment of the present invention;

[0025] Figure 2 It is a flowchart of another short-term solar radiation intensity prediction method provided by an embodiment of the present invention;

[0026] Figure 3 Flow chart of yet another short - term solar radiation intensity prediction method provided by an embodiment of the present invention;

[0027] Figure 4 Comparison diagram of predicting and actually observing the surface radiation of 4 stations in a target area within a certain period provided by an embodiment of the present invention;

[0028] Figure 5 Another comparison diagram of predicting and actually observing the surface radiation of 4 stations in a target area within a certain period provided by an embodiment of the present invention;

[0029] Figure 6 Structure block diagram of a short - term solar radiation intensity prediction device provided by an embodiment of the present invention;

[0030] Figure 7 Structure diagram of an electronic device provided by an embodiment of the present invention. Detailed implementation manners

[0031] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0032] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above - mentioned drawings are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such used data can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order different from those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non - exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily have to be limited to those clearly listed steps or units, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0033] In the technical solution of this application, the acquisition, storage, use, processing, etc. of data all comply with relevant regulations.

[0034] In one embodiment, Figure 1The flowchart of a short-term solar radiation intensity prediction method provided by an embodiment of the present invention. This embodiment is applicable to the situation of predicting the short-term solar radiation intensity in a specified target area. This method can be executed by a short-term solar radiation intensity prediction device, which can be implemented in the form of hardware and / or software, and can be configured in an electronic device.

[0035] As Figure 1 shown, the specific steps of the short-term solar radiation intensity prediction method in this embodiment include:

[0036] S110. Obtain the historical real solar radiation data corresponding to each ground station in the target area during the historical preset time period, and obtain the historical meteorological element reanalysis data and historical satellite observation data during the historical preset time period from at least two third-party databases.

[0037] Among them, the target area is the area selected by the user that needs to conduct short-term solar radiation intensity prediction, and this area is selected by itself according to requirements. The historical preset time period refers to the range of historical time periods selected according to user needs. The historical real solar radiation data is the real solar radiation data during the historical preset time period, which can be understood as the real solar radiation observation data respectively collected by each ground station. The time resolution of this solar radiation observation data can be 1 hour, indicating that the time interval of the data or observation is once per hour. The real solar radiation observation data can include global horizontal irradiance (GHI), diffuse horizontal irradiance (DHI), and direct normal irradiance (DNI).

[0038] In this embodiment, the third-party database can be understood as some publicly available databases storing historical meteorological element reanalysis data and historical satellite observation data. The third-party databases in this embodiment can include but are not limited to the European Centre for Medium-Range Weather Forecasts (ECMWF) database and the second-generation geostationary meteorological satellite database. In this embodiment, the historical meteorological element reanalysis data during the historical preset time period can be obtained from the ECMWF database, and the historical satellite observation data during the historical preset time period can be obtained from the second-generation geostationary meteorological satellite database. It should be noted that the acquisition time periods corresponding to the historical real solar radiation data, historical meteorological element reanalysis data, and historical satellite observation data in this embodiment are the same time period.

[0039] In this embodiment, the target area may include multiple ground stations, each corresponding to respective location information. The location information of each ground station includes at least: longitude and latitude information and altitude information. Obtain the historical true solar radiation data corresponding to each ground station in the target area during a historical preset time period, and obtain the historical meteorological element reanalysis data and historical satellite observation data from multiple third-party databases during the same historical preset time period according to requirements; among them, the historical meteorological element reanalysis data may include, but is not limited to, temperature, humidity, wind speed, air pressure, cloud amount, etc.; the historical satellite observation data may include, but is not limited to, ground radiation intensity, cloud cover, aerosol concentration, ground reflectivity, and information such as the solar angle and satellite angle during satellite observation. The satellite angle includes the satellite zenith angle and the satellite azimuth angle, and the satellite zenith angle affects the satellite's observation range. In this embodiment, the spatial resolution corresponding to the historical meteorological element reanalysis data may be 25 km, indicating that each data point covers an area of 25 km × 25 km; the time resolution is 1 hour, indicating that the time interval of the data or observation is once per hour. Similarly, the same is true for the historical satellite observation data. Of course, the selection of the spatial resolution and time resolution can also be selected according to requirements, and this embodiment does not limit it here.

[0040] S120. Among the historical meteorological element reanalysis data and the historical satellite observation data, respectively match the target reanalysis data and the target satellite observation data corresponding to the ground station based on the location information of the ground station.

[0041] Among them, the target reanalysis data is the reanalysis data selected from the historical meteorological element reanalysis data that has the same time and the same location as the historical true solar radiation data. Similarly, the target satellite observation data is the satellite observation data selected from the historical satellite observation data that has the same time and the same location as the historical true solar radiation data.

[0042] In this embodiment, it is necessary to respectively match, among the historical meteorological element reanalysis data and the historical satellite observation data, the reanalysis data and the satellite observation data that have the same time and the same location as the historical true solar radiation data. It can be understood that, on the premise of ensuring the unified data format, ensure that the time resolutions of the two types of data are consistent in time (such as hours, days, months), and ensure that the spatial resolutions of the two types of data are consistent in space (such as longitude and latitude grids). If the time resolutions of the two types of data are different, they can be aligned by interpolation or aggregation methods. If the spatial resolutions of the two types of data are different, the high-resolution data can be resampled to the low-resolution grid, or the low-resolution data can be interpolated to the high-resolution grid.

[0043] In some embodiments, the historical timestamps corresponding to the historical true solar radiation data can be converted to UTC time to obtain the converted target historical true solar radiation data, so as to unify the time of the historical true solar radiation data, the historical meteorological element reanalysis data, and the historical satellite observation data. Then, for the target historical true solar radiation data after time unification, a site location grid is formed based on the location information of each ground site, and a spatio-temporal matching method is used to extract the target reanalysis data corresponding to each pixel in the site location grid from the historical meteorological element reanalysis data, and the target satellite observation data corresponding to each pixel in the site location grid is extracted from the historical satellite observation data.

[0044] S130. Train a short-term radiation prediction model based on the target reanalysis data, the target satellite observation data, and the historical true solar radiation data to obtain a trained short-term radiation prediction model; wherein, the short-term radiation prediction model is composed of an XGBoost model.

[0045] Among them, the XGBoost model is a strong prediction model constructed by at least two decision trees.

[0046] In this embodiment, the target reanalysis data, the target satellite observation data, and the historical true solar radiation data can be used as a training set to train the short-term radiation prediction model, so as to obtain a trained short-term radiation prediction model. Specifically, the historical true solar radiation data is used as a label, and the target reanalysis data and the target satellite observation data are used as feature values to construct a sample set; the data in the sample set is screened and converted for solar angles, screened and converted for satellite angles, and the normalized difference vegetation index NDVI is calculated. Then, it is input into the short-term radiation prediction model, and the preset Bayesian hyperparameter optimization method, genetic algorithm, and gradient optimization method are used to optimize the parameters of the model, so as to optimize the hyperparameters of the short-term radiation prediction model until the loss function reaches the minimum, and a trained short-term radiation prediction model is obtained. In this embodiment, in addition to being based on XGBoost, the short-term radiation prediction model can also be a random forest or other neural network models, which are not limited in this embodiment.

[0047] S140. Obtain the current surface meteorological element data and the current satellite observation data, and predict the solar radiation intensity of the ground site in the short term in the future based on the current surface meteorological element data, the current satellite observation data, and the trained short-term radiation prediction model to obtain a prediction result.

[0048] Among them, the current surface meteorological element data refers to the surface meteorological element data collected in real time currently, and the surface meteorological element data may include temperature, humidity, wind speed, air pressure, cloud amount, etc.; the current satellite observation data refers to the satellite observation data collected in real time currently, and the satellite observation data may include information such as ground radiation intensity, cloud cover, aerosol concentration, ground reflectivity, and solar angle and satellite angle during satellite observation.

[0049] In this embodiment, the current surface meteorological element data is obtained from the meteorological element collection points of each ground station, and the current satellite observation data is obtained from the satellite collection points. Then, the collected current surface meteorological element data and current satellite observation data are respectively processed, and the processed current surface meteorological element data and current satellite observation data are input into the trained short-term radiation prediction model to predict the solar radiation intensity of the ground station in the short term in the future to obtain a prediction result. The short term in this embodiment can be the prediction of the solar radiation intensity within 0-4 hours.

[0050] The technical solution of the embodiment of the present invention, by fusing historical meteorological element reanalysis data and historical satellite observation data, in the historical meteorological element reanalysis data and historical satellite observation data, respectively match the target reanalysis data and target satellite observation data corresponding to the ground station based on the position information of the ground station. Then, the short-term radiation prediction model is trained by the target reanalysis data, target satellite observation data, and historical real solar radiation data to obtain a trained short-term radiation prediction model. On this basis, based on the current surface meteorological element data, current satellite observation data, and trained short-term radiation prediction model, the solar radiation intensity of the ground station in the short term in the future is predicted to obtain a prediction result, which can improve the prediction accuracy and accuracy of the solar radiation intensity in the short term. Especially in the case of cloud occlusion and sudden weather changes, it can provide more accurate prediction results, so as to accurately predict the corresponding solar energy resources, better help the solar power station to make reasonable scheduling, and improve the power generation efficiency.

[0051] In one embodiment, the method further includes:

[0052] Visualize the prediction result so that the user can determine the solar radiation intensity, solar energy resource distribution, and power generation potential in the future period according to the visualized prediction result to assist the power station at the ground station in making scheduling decisions and operation optimization;

[0053] Among them, the visualization methods at least include: scatter plot visualization and time series plot visualization.

[0054] In this embodiment, the prediction results are visually displayed in the form of charts, time-series line charts, etc. Users can, based on these results, understand information such as solar radiation intensity, solar energy resource distribution, and power generation potential in a future period of time, to assist the solar power station in making scheduling decisions and optimizing operations. Specifically, the scatter plot visualization is to plot the predicted and actual results in a scatter plot, and calculate error metrics such as the correlation coefficient R and the root mean square error RMSE, which can determine the deviation degree between the prediction and the actual results and can be used as a reference basis for different stations. It can be understood that the irradiance is related to the power generation of the station, and the power generation of this station can be known according to the predicted results. The time-series chart visualization is to plot the predicted and actual results as line charts with time as the x-axis according to different stations, which can determine the deviation degree between the prediction and the actual results in different scenarios and can be used as a reference basis for the characteristics of different stations.

[0055] In one embodiment, Figure 2 is a flowchart of another short-term solar radiation intensity prediction method provided by an embodiment of the present invention. Based on the above embodiments, in the historical meteorological element reanalysis data and historical satellite observation data, the target reanalysis data and target satellite observation data corresponding to the ground station are respectively matched based on the position information of the ground station; the short-term radiation prediction model is trained based on the target reanalysis data, target satellite observation data, and historical true solar radiation data to obtain a trained short-term radiation prediction model; the solar radiation intensity of the ground station in the short-term future is predicted based on the current surface meteorological element data, current satellite observation data, and the trained short-term radiation prediction model to obtain prediction results, which is further refined.

[0056] As Figure 2 shown, the short-term solar radiation intensity prediction method in this embodiment may specifically include the following steps:

[0057] S210. Obtain the historical true solar radiation data corresponding to each ground station in the target area during the historical preset time period, and obtain the historical meteorological element reanalysis data and historical satellite observation data during the historical preset time period from at least two third-party databases.

[0058] S220. Preprocess the historical true solar radiation data, and perform UTC time conversion on the historical timestamps corresponding to the preprocessed historical true solar radiation data to obtain the converted target historical true solar radiation data, so as to unify the time of the historical true solar radiation data, historical meteorological element reanalysis data, and historical satellite observation data.

[0059] Among them, the target historical true solar radiation data is the historical true solar radiation data after UTC time conversion. The UTC time conversion in this embodiment is to unify the time.

[0060] In this embodiment, the methods for preprocessing historical real solar radiation data may include, but are not limited to, data cleaning, handling missing values, outliers, time alignment, spatial interpolation, etc. After processing, perform UTC time conversion on the historical timestamps corresponding to the preprocessed historical real solar radiation data to obtain the converted target historical real solar radiation data, so as to unify the time of the historical real solar radiation data with the historical meteorological element reanalysis data and the historical satellite observation data.

[0061] S230. For the target historical real solar radiation data after time unification, form a site location grid based on the location information of each ground site; wherein, the site location grid includes pixels; each pixel represents a ground site respectively.

[0062] In this embodiment, for the target historical real solar radiation data after time unification, form a site location grid based on the location information of each ground site; wherein, the site location grid includes pixels; each pixel represents a ground site respectively. In this embodiment, the site location grid is used for spatial analysis and visualization of the site locations. Each grid cell (pixel) represents a specific geographical area, and the site locations can be assigned to the corresponding grid cells according to their longitude and latitude coordinates.

[0063] S240. Extract the target reanalysis data corresponding to each pixel in the site location grid from the historical meteorological element reanalysis data by using a spatio-temporal matching method; wherein, the target reanalysis data is consistent with the target historical real solar radiation data in terms of time and space.

[0064] Among them, the spatio-temporal matching method is a method for performing time and space matching.

[0065] In this embodiment, extract the target reanalysis data corresponding to each pixel in the site location grid from the historical meteorological element reanalysis data by using a spatio-temporal matching method; wherein, the target reanalysis data is consistent with the target historical real solar radiation data in terms of time and space. Specifically, extract the time records in the historical meteorological element reanalysis data that have the same timestamp as the historical solar radiation intensity, and then perform spatial location matching to obtain the corresponding target reanalysis data.

[0066] S250. Extract the target satellite observation data corresponding to each pixel in the site location grid from the historical satellite observation data by using a spatio-temporal matching method; wherein, the target satellite observation data is consistent with the target historical real solar radiation data in terms of time and space.

[0067] In this embodiment, a spatio-temporal matching method is adopted to extract the target satellite observation data corresponding to each pixel in the site location grid from historical satellite observation data; wherein, the time and space of the target satellite observation data are consistent with those of the target historical true solar radiation data. Specifically, time records with the same timestamp as the historical solar radiation intensity are extracted from the historical satellite observation data, and then spatial position matching is performed to obtain the corresponding target reanalysis data.

[0068] S260. Use the historical true solar radiation data as a label, and use the target reanalysis data and the target satellite observation data as eigenvalue to construct a sample set.

[0069] In this embodiment, the historical true solar radiation data is used as a label, and the target reanalysis data and the target satellite observation data are used as eigenvalue to construct a sample set.

[0070] S270. Perform data preprocessing on the sample set to obtain a processed target sample set; wherein, the data preprocessing methods at least include: screening and conversion of solar angles, screening and conversion of satellite angles, and calculation of the normalized difference vegetation index NDVI.

[0071] In this embodiment, data preprocessing is performed on the sample set to obtain a processed target sample set; wherein, the data preprocessing methods at least include: screening and conversion of solar angles, screening and conversion of satellite angles, and calculation of the normalized difference vegetation index NDVI. Among them, NDVI = (NIR - Red) ÷ (NIR + Red), and the normalized difference vegetation index (NDVI) method is a normalized index used to generate images showing the amount of vegetation (relative biomass). This index compares the characteristics of two bands in a multispectral raster dataset, i.e., the pigment absorption rate of chlorophyll in the red light band and the high reflectivity of plant bodies in the NIR band. NIR = pixel value of the near-infrared band; Red = pixel value of the red light band. The output value of this index ranges from -1 to 1.

[0072] In one embodiment, the solar angles include: solar zenith angle and solar azimuth angle; the satellite angles include: satellite zenith angle and satellite altitude angle; wherein, the cosine conversion of the solar zenith angle is: cos_solar_zenith_angle = cos(solar_zenith_angle × π ÷ 180); the cosine conversion of the solar azimuth angle is: cos_solar_altitude_angle = cos(solar_altitude_angle × π ÷ 180); the cosine conversion of the satellite zenith angle is: cos_satellite_zenith_angle = cos(satellite_zenith_angle × π ÷ 180); the cosine conversion of the satellite altitude angle is: cos_satellite_altitude_angle = cos(satellite_altitude_angle × π ÷ 180).

[0073] S280. Input the target sample set into the short-term radiation prediction model, and use the preset Bayesian hyperparameter optimization method to optimize the hyperparameters of the short-term radiation prediction model until the loss function reaches the minimum, thereby obtaining the trained short-term radiation prediction model.

[0074] Among them, the minimum of the loss function indicates that the difference between the predicted value and the true value is the smallest; the hyperparameters at least include: the maximum depth of the tree, the learning rate, the subsample ratio, the column sampling ratio, the regularization parameter, and the number of trees.

[0075] The preset Bayesian hyperparameter optimization method mentioned in this embodiment is an optimization method based on Bayesian statistics, which is used to efficiently search for the best hyperparameter combination of a machine learning model. Bayesian optimization guides the search process by constructing a probability model of the objective function, so as to find a better hyperparameter combination within fewer iterations.

[0076] In this embodiment, when training the prediction model, the training data set can be divided into a training set and a test set according to the ratio of 80% and 20%. By parameter search, the hyperparameters of the xgboost algorithm are determined, the model is trained on the training set, and the test set is used to test the model to check whether problems such as overfitting and underfitting occur in the model, so as to ensure the generalization of the model.

[0077] In this embodiment, the Bayesian hyperparameter optimization method is used to optimize hyperparameters such as the maximum depth of the tree (max_depth), the learning rate (learning_rate), the subsample ratio (subsample), and the column sampling ratio (colsample_bytree). Specifically, the learning rate (Learning Rate): controls the learning speed of the model; the maximum depth of the tree (MaxDepth): the maximum depth of the decision tree; the subsample ratio (Subsample): the sample ratio used for each tree; the feature sampling ratio (Colsample_bytree): the feature ratio used for each tree; the regularization parameter (Lambda, Alpha): controls the complexity of the model; the number of trees (N_estimators): the number of decision trees. The specific steps are as follows: First, set the Bayesian optimization parameter space, and then define the specific process of the optimization objective function, including defining the objective function to be optimized and the alternative parameter space, etc. Among them, the objective function of XGBoost includes two parts: the loss function: measures the difference between the predicted value and the true value of the model; the regularization term: controls the complexity of the model and prevents overfitting. In each step, a new decision tree is added to fit the residual (i.e., the negative gradient) of the current model, and the model parameters are updated by the gradient descent method. The input of the objective function is the hyperparameter combination, and the output is the performance of the model on the validation set.

[0078] S290. Input the current surface meteorological element data and the current satellite observation data into the trained short-term radiation prediction model to obtain the prediction result of the solar radiation intensity at the ground station in the short term in the future, so as to determine the future power generation of the ground station according to the prediction result.

[0079] In this embodiment, the current surface meteorological element data and the current satellite observation data are processed, and then the processed current surface meteorological element data and the current satellite observation data are input into the trained short-term radiation prediction model to obtain the prediction result of the solar radiation intensity at the ground station in the short term in the future, so as to determine the future power generation of the ground station according to the prediction result.

[0080] In the above technical solution of this embodiment, for the target historical true solar radiation data after time unification, a site location grid is formed based on the location information of each ground station, and each pixel represents a ground station. The space-time matching method is used to extract the target reanalysis data corresponding to each pixel in the site location grid from the historical meteorological element reanalysis data, and the space-time matching method is used to extract the target satellite observation data corresponding to each pixel in the site location grid from the historical satellite observation data, which can accurately match the data in time and space and provide a basis for subsequent processing; by screening and converting the solar angle, screening and converting the satellite angle, and calculating the normalized difference vegetation index NDVI for the sample set, a processed target sample set is obtained. Based on this, the target sample set is input into the short-term radiation prediction model, and the hyperparameters of the short-term radiation prediction model are optimized using a preset Bayesian hyperparameter optimization method until the loss function reaches the minimum, and a trained short-term radiation prediction model is obtained, so as to predict the solar radiation intensity at the ground station in the short term in the future, and determine the future power generation of the ground station according to the prediction result. Through machine learning and deep learning algorithms, this model can adapt to different regions and climate conditions, realize flexible solar radiation forecasting, and can be extended to the solar energy resource assessment in various regions. On the basis of the above embodiment, this embodiment can further improve the prediction accuracy and accuracy of the solar radiation intensity in the short term, especially in the case of cloud occlusion and sudden weather changes, and can provide more accurate prediction results, so as to accurately predict the corresponding solar energy resources and better help the solar power station to carry out reasonable scheduling and improve the power generation efficiency.

[0081] In one embodiment, for better understanding of the short-term prediction of solar radiation intensity, Figure 3 FIG. is a flowchart of another short-term solar radiation intensity prediction method provided by an embodiment of the present invention. Figure 4 FIG. is a comparison diagram of predicting and actually observing the surface radiation of 4 stations in the target area within a certain time period provided by an embodiment of the present invention. Figure 5Another schematic diagram of comparing the prediction and actual observation of surface radiation of four stations in a target area within a certain time period provided by an embodiment of the present invention, in this embodiment, Figure 4 The test duration is 1 hour. Figure 5 Taking the prediction time of 4 hours as an example, it can be seen that the generalization is good in the predictions of different stations, there is no specific deviation, and the prediction results are highly consistent with the actual observations. The prediction accuracy is greatly improved compared with the traditional algorithm. Figure 4 and Figure 5 In it, R represents the correlation coefficient (that is, the correlation between the prediction and the actual), RMSE represents the root mean square error, y represents the equation of the line, and N represents the number of data points at the predicted time.

[0082] like Figure 3 As shown, specifically, the specific steps of the short-term prediction method of solar radiation intensity are as follows:

[0083] a1. Obtain historical real solar radiation data, select the study area, and obtain the minute irradiance data of the meteorological radiation station in the study area in the preset time period (for example, it can be the whole year of 2018), including global radiation irradiance (GHI), diffuse radiation irradiance (DHI), and direct radiation irradiance (DNI); remove the data with null values ​​and convert the local time to UTC time. Each meteorological radiation station contains the station information of latitude, longitude and altitude.

[0084] a2. Download meteorological element reanalysis data from the first third party and obtain meteorological element reanalysis data for the same preset time period as the study area (e.g., global data for 2018), with a spatial resolution of 25 km and a temporal resolution of 1 hour. Perform spatial interpolation on the reanalysis data to increase the spatial resolution to 4 km, which is consistent with the satellite input.

[0085] a3. Download satellite observation data from a second third party. Similarly, the satellite observation data for the same preset time period as the study area (e.g. global data for 2018) with a temporal resolution of 1 hour and a spatial resolution of 4 km.

[0086] a4. Data preprocessing and fusion module. Perform spatial matching on satellite radiation data, meteorological data and surface radiation data, and align the temporal and spatial differences of different data sources. Fuse multi-source data to form a training data set. Fuse historical solar radiation data, meteorological element reanalysis data, and satellite observation data into a data set for the whole year of 2018, and perform corresponding preprocessing. Use surface radiation observation data as labels, reanalysis data, and satellite observation data as feature values ​​to merge into a training data set, and perform training data set cleaning, including solar angle screening, cosine transformation, etc.

[0087] a5. Short-term solar radiation prediction model. By integrating the machine learning algorithm Xgboost, it is trained on the training data set formed by remote sensing data, meteorological data and surface solar radiation data, and the generalization of the model is tested through the test set to establish a solar radiation prediction model for the Jiangsu region. The model can predict the surface solar radiation intensity and change trend within the next 0-4 hours.

[0088] a6. Forecast result output and visualization module. The prediction results are visually displayed in the form of charts, time series line charts, etc. Users can understand information such as the solar radiation intensity, solar energy resource distribution, and power generation potential within a certain period in the future based on these results, which helps the solar power station to make scheduling decisions and optimize operations.

[0089] In one embodiment, Figure 6 The following is a structural block diagram of a short-term solar radiation intensity prediction device provided by an embodiment of the present invention. This device is applicable to the situation of predicting the short-term solar radiation intensity of a specified target area, and this device can be implemented by hardware / software. It can be configured in an electronic device to implement a short-term solar radiation intensity prediction method in an embodiment of the present invention.

[0090] As Figure 6 shown, the device includes: a data acquisition module 610, a matching module 620, a model training module 630, and a prediction module 640;

[0091] Among them, the data acquisition module 610 is used to obtain the historical true solar radiation data corresponding to each ground station in the target area during the historical preset time period, and obtain the historical meteorological element reanalysis data and historical satellite observation data during the historical preset time period from at least two third-party databases;

[0092] The matching module 620 is used to respectively match the target reanalysis data and the target satellite observation data corresponding to the ground station based on the position information of the ground station in the historical meteorological element reanalysis data and the historical satellite observation data;

[0093] The model training module 630 is used to train the short-term radiation prediction model based on the target reanalysis data, the target satellite observation data, and the historical true solar radiation data to obtain a trained short-term radiation prediction model; among them, the short-term radiation prediction model is composed of an XGBoost model;

[0094] A prediction module 640, configured to obtain current surface meteorological element data and current satellite observation data, and predict the solar radiation intensity of the ground station in the short term in the future based on the current surface meteorological element data, the current satellite observation data, and the trained short-term radiation prediction model to obtain a prediction result.

[0095] In an embodiment of the present invention, a matching module, by respectively matching the target reanalysis data and the target satellite observation data corresponding to the ground station based on the position information of the ground station in the historical meteorological element reanalysis data and the historical satellite observation data, and a model training module, training the short-term radiation prediction model through the target reanalysis data, the target satellite observation data, and the historical true solar radiation data to obtain a trained short-term radiation prediction model. On this basis, the prediction module predicts the solar radiation intensity of the ground station in the short term in the future based on the current surface meteorological element data, the current satellite observation data, and the trained short-term radiation prediction model to obtain a prediction result, which can improve the prediction accuracy and accuracy of the solar radiation intensity in the short term. Especially in the case of cloud cover and sudden weather changes, it can provide more accurate prediction results, so as to accurately predict the corresponding solar energy resources, better help the solar power station to carry out reasonable scheduling, and improve the power generation efficiency.

[0096] In an embodiment, the position information of the ground station at least includes: longitude and latitude information and altitude information; correspondingly, the matching module 620 includes:

[0097] A data processing unit, configured to preprocess the historical true solar radiation data, and perform UTC time conversion on the historical timestamp corresponding to the preprocessed historical true solar radiation data to obtain the converted target historical true solar radiation data, so as to unify the time of the historical true solar radiation data with the historical meteorological element reanalysis data and the historical satellite observation data;

[0098] A grid formation unit, configured to form a site position grid according to the position information of each ground station for the target historical true solar radiation data after time unification; wherein, the site position grid includes pixels; each pixel respectively represents a ground station;

[0099] A first data extraction unit, configured to extract the target reanalysis data corresponding to each pixel in the site position grid from the historical meteorological element reanalysis data by using a spatio-temporal matching method; wherein, the target reanalysis data is consistent with the target historical true solar radiation data in terms of time and space;

[0100] A second data extraction unit, configured to extract the target satellite observation data corresponding to each pixel in the site location grid from the historical satellite observation data by using the spatio-temporal matching method; wherein, the time and space of the target satellite observation data are consistent with those of the target historical true solar radiation data.

[0101] In one embodiment, the XGBoost model is a strong prediction model constructed by at least two decision trees; correspondingly, the model training module 630 includes:

[0102] A sample set construction unit, configured to use the historical true solar radiation data as labels, and use the target reanalysis data and the target satellite observation data as feature values to construct a sample set;

[0103] A processing unit, configured to perform data preprocessing on the sample set to obtain a processed target sample set; wherein, the data preprocessing methods at least include: screening and conversion of solar angles, screening and conversion of satellite angles, and calculation of the normalized difference vegetation index NDVI;

[0104] A training unit, configured to input the target sample set into the short-term radiation prediction model, and use a preset Bayesian hyperparameter optimization method to optimize the hyperparameters of the short-term radiation prediction model until the loss function reaches the minimum, so as to obtain a trained short-term radiation prediction model; wherein, the minimum of the loss function indicates that the difference between the predicted value and the true value is the smallest; the hyperparameters at least include: the maximum depth of the tree, the learning rate, the subsample ratio, the column sampling ratio, the regularization parameter, and the number of trees.

[0105] In one embodiment, the solar angles include: solar zenith angle and solar azimuth angle; the satellite angles include: satellite zenith angle and satellite altitude angle;

[0106] Wherein, the cosine conversion of the solar zenith angle is: cos_solar_zenith_angle = cos(solar_zenith_angle × π ÷ 180); the cosine conversion of the solar azimuth angle is: cos_solar_altitude_angle = cos(solar_altitude_angle × π ÷ 180); the cosine conversion of the satellite zenith angle is: cos_satellite_zenith_angle = cos(satellite_zenith_angle × π ÷ 180); the cosine conversion of the satellite altitude angle is: cos_satellite_altitude_angle = cos(satellite_altitude_angle ×

[0107] π ÷ 180).

[0108] In one embodiment, the prediction module 640 includes:

[0109] A prediction unit, configured to input the current surface meteorological element data and the current satellite observation data into the trained short-term radiation prediction model, so as to obtain a prediction result of the solar radiation intensity of the ground station in the short term in the future, and determine the future power generation of the ground station according to the prediction result.

[0110] In one embodiment, the device further includes:

[0111] A display module, configured to visually display the prediction result, so that a user can determine the solar radiation intensity, solar energy resource distribution, and power generation potential in a future period according to the visually displayed prediction result, to assist the power station of the ground station in making scheduling decisions and operation optimization;

[0112] Wherein, the way of the visual display at least includes: scatter plot visualization and time series plot visualization.

[0113] The short-term solar radiation intensity prediction device provided by the embodiments of the present invention can execute the short-term solar radiation intensity prediction method provided by any embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the method.

[0114] In one embodiment, Figure 7 It is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as, laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are only examples, and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0115] As Figure 7 shown, the electronic device 10 includes at least one processor 11, and a memory communicatively connected to at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc. Among them, the memory stores a computer program executable by at least one processor, and the processor 11 can execute various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. The input / output (I / O) interface 15 is also connected to the bus 14.

[0116] Multiple components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disc, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0117] The processor 11 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the short-term prediction method of solar radiation intensity.

[0118] In some embodiments, the short-term prediction method of solar radiation intensity can be implemented as a computer program, which is tangibly included in a computer-readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the short-term prediction method of solar radiation intensity described above can be executed. Alternatively, in other embodiments, the processor 11 can be configured to execute the short-term prediction method of solar radiation intensity by any other suitable means (e.g., by means of firmware).

[0119] The various embodiments of the systems and technologies described above in this article can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on a chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs, the one or more computer programs can be executed and / or interpreted on a programmable system including at least one programmable processor, the programmable processor can be a dedicated or general-purpose programmable processor, can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0120] A computer program for implementing the method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to the processors of general-purpose computers, special-purpose computers, or other programmable solar radiation intensity short-term prediction devices, such that when executed by the processors, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The computer programs can be executed entirely on the machine, partially on the machine, as a stand-alone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0121] In the context of the present invention, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium can be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0122] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0123] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected with each other by digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), blockchain network, and the Internet.

[0124] A computing system can include a client and a server. The client and the server are generally remote from each other and typically interact via a communication network. The client-server relationship is created by computer programs that run on respective computers and have a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system and solves the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services.

[0125] It should be understood that various forms of the processes shown above can be used, steps can be reordered, added, or deleted. For example, the steps recited in the present invention can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is made herein.

[0126] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A short-term prediction method for solar radiation intensity, characterized in that, The method includes: Obtaining the historical true solar radiation data corresponding to each ground station in the target area during a historical preset time period, and obtaining the historical meteorological element reanalysis data and historical satellite observation data during the historical preset time period from at least two third-party databases; Among the historical meteorological element reanalysis data and the historical satellite observation data, respectively, based on the position information of the ground station, matching the target reanalysis data and the target satellite observation data corresponding to the ground station; Training a short-term radiation prediction model based on the target reanalysis data, the target satellite observation data, and the historical true solar radiation data to obtain a trained short-term radiation prediction model; wherein, the short-term radiation prediction model is composed of an XGBoost model; Obtaining the current surface meteorological element data and the current satellite observation data, and predicting the solar radiation intensity of the ground station in the short term in the future based on the current surface meteorological element data, the current satellite observation data, and the trained short-term radiation prediction model to obtain a prediction result.

2. The method according to claim 1, characterized in that The position information of the ground station includes at least: longitude and latitude information and altitude information; Correspondingly, the step of, among the historical meteorological element reanalysis data and the historical satellite observation data, respectively, based on the position information of the ground station, matching the target reanalysis data and the target satellite observation data corresponding to the ground station includes: Preprocessing the historical true solar radiation data, and performing UTC time conversion on the historical timestamp corresponding to the preprocessed historical true solar radiation data to obtain the converted target historical true solar radiation data, so as to unify the time of the historical true solar radiation data with the historical meteorological element reanalysis data and the historical satellite observation data; For the target historical true solar radiation data after time unification, forming a site position grid according to the position information of each ground station; wherein, the site position grid includes pixels; each pixel represents a ground station; Extracting the target reanalysis data corresponding to each pixel in the site position grid from the historical meteorological element reanalysis data by using a spatio-temporal matching method; wherein, the target reanalysis data is consistent with the target historical true solar radiation data in time and space; Extracting the target satellite observation data corresponding to each pixel in the site position grid from the historical satellite observation data by using the spatio-temporal matching method; wherein, the target satellite observation data is consistent with the target historical true solar radiation data in time and space.

3. The method according to claim 1, characterized in that The XGBoost model is a strong prediction model constructed by at least two decision trees; correspondingly, the step of training a short-term radiation prediction model based on the target reanalysis data, the target satellite observation data, and the historical true solar radiation data to obtain a trained short-term radiation prediction model includes: Taking the historical true solar radiation data as a label, and taking the target reanalysis data and the target satellite observation data as feature values to construct a sample set; Perform data preprocessing on the sample set to obtain a processed target sample set; wherein, the data preprocessing methods at least include: screening and conversion of solar angles, screening and conversion of satellite angles, and calculation of the Normalized Difference Vegetation Index (NDVI). Input the target sample set into the short-term radiation prediction model, and use a preset Bayesian hyperparameter optimization method to optimize the hyperparameters of the short-term radiation prediction model until the loss function reaches the minimum, obtaining a trained short-term radiation prediction model; wherein, the minimum of the loss function indicates that the difference between the predicted value and the true value is the smallest; the hyperparameters at least include: the maximum depth of the tree, the learning rate, the subsample ratio, the column sampling ratio, the regularization parameter, and the number of trees.

4. The method according to claim 3, wherein The solar angles include: solar zenith angle and solar azimuth angle; the satellite angles include: satellite zenith angle and satellite altitude angle. Among them, the cosine conversion of the solar zenith angle is: cos_solar_zenith_angle = cos(solar_zenith_angle × π ÷ 180); the cosine conversion of the solar azimuth angle is: cos_solar_altitude_angle = cos(solar_altitude_angle × π ÷ 180); the cosine conversion of the satellite zenith angle is: cos_satellite_zenith_angle = cos(satellite_zenith_angle × π ÷ 180); the cosine conversion of the satellite altitude angle is: cos_satellite_altitude_angle = cos(satellite_altitude_angle × π ÷ 180).

5. The method according to claim 1, wherein Predicting the solar radiation intensity of the ground station in the short term in the future based on the current surface meteorological element data, the current satellite observation data, and the trained short-term radiation prediction model to obtain a prediction result, including:[[]] Input the current surface meteorological element data and the current satellite observation data into the trained short-term radiation prediction model to obtain a prediction result of the solar radiation intensity of the ground station in the short term in the future, so as to determine the future power generation of the ground station according to the prediction result.

6. The method according to claim 1, wherein The method further includes:[[]] Visualize the prediction result so that users can determine the solar radiation intensity, solar energy resource distribution, and power generation potential within a certain period in the future according to the visualized prediction result, to assist the power station of the ground station in making scheduling decisions and operation optimization. Among them, the visualization methods at least include: scatter plot visualization and time series plot visualization.

7. A short-term prediction device for solar radiation intensity, characterized in that, The device includes:[[]] A data acquisition module, configured to acquire the historical true solar radiation data corresponding to each ground station in the target area during a historical preset period, and acquire the historical meteorological element reanalysis data and historical satellite observation data during the historical preset period from at least two third-party databases. A matching module, configured to respectively match the target reanalysis data and the target satellite observation data corresponding to the ground station based on the location information of the ground station in the historical meteorological element reanalysis data and the historical satellite observation data. A model training module, configured to train a short-term radiation prediction model based on the target reanalysis data, the target satellite observation data, and the historical true solar radiation data, so as to obtain a trained short-term radiation prediction model; wherein, the short-term radiation prediction model is constituted by an XGBoost model; A prediction module, configured to obtain current surface meteorological element data and current satellite observation data, and predict the solar radiation intensity of the ground site in the short term in the future based on the current surface meteorological element data, the current satellite observation data, and the trained short-term radiation prediction model to obtain a prediction result.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor, so that the at least one processor can execute the short-term solar radiation intensity prediction method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions, and the computer instructions are used to implement the short-term solar radiation intensity prediction method according to any one of claims 1-6 when executed by a processor.

10. A computer program product, characterized in that, The computer program product includes a computer program, and the computer program implements the short-term solar radiation intensity prediction method according to any one of claims 1-6 when executed by a processor.

Images