Meteorological data prediction method and system based on micrometeorological model
By constructing a meteorological data prediction method based on micrometeorological model, the problems of result distortion and difficulty in selecting parameters in the construction of meteorological models in small-scale areas are solved, and more accurate and convenient meteorological prediction is achieved.
Patent Information
- Application Number
- CN202510217797.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-26
- Publication Date
- 2025-05-30
AI Technical Summary
The existing meteorological interpolation model has problems such as distortion of results, misleading predictions and difficulty in selecting parameters when constructing meteorological models of small scale areas, especially in small areas with severe terrain changes.
The meteorological data prediction method based on the micrometeorological model is adopted, and the coordinate data of the meteorological stations to be predicted and the actual meteorological values of each meteorological station in the historical period are obtained, and the neural network model is trained to construct a micrometeorological model for prediction.
The construction of micrometeorological field models in small-scale areas is realized, making meteorological predictions in small-scale areas more accurate and convenient, avoiding the problem of result distortion and difficulty in parameter selection in traditional methods.
Smart Images
Figure CN120065381A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of meteorological prediction technology, and particularly to a meteorological data prediction method and system based on a micrometeorological model. Background Art
[0002] The construction of meteorological models is of great significance in meteorology. Through meteorological models, based on limited meteorological observation data, the numerical values of meteorological elements at other locations and times can be inferred in space and time, thus filling in the data gaps and obtaining more comprehensive and continuous meteorological information. This is crucial for meteorological forecasting, climate research, disaster warning, etc. Through meteorological models, meteorological changes can be analyzed and predicted more accurately, improving the efficiency and accuracy of meteorological services. The micrometeorological model is for the simulation and prediction of small-scale meteorological phenomena, which is of great significance for in-depth understanding of small-scale meteorological phenomena and predicting sudden meteorological events in specific regions.
[0003] Currently, the commonly used method for constructing meteorological interpolation models is Kriging interpolation. The Kriging interpolation method is not suitable for constructing small-area meteorological models with non-linear changes. This method has limited effects in small areas with drastic terrain changes, which may lead to problems such as distorted results and misleading predictions, and it is difficult to select parameters, requiring multiple trials and errors. Therefore, there is an urgent need to propose a solution that can accurately predict meteorology in small-scale regions. Summary of the Invention
[0004] This application provides a meteorological data prediction method and system based on a micrometeorological model to at least solve the technical problems that it has limited effects in small areas with drastic terrain changes, may lead to problems such as distorted results and misleading predictions, and it is difficult to select parameters, requiring multiple trials and errors.
[0005] The first aspect of the embodiments of this application proposes a meteorological data prediction method based on a micrometeorological model. The method includes:
[0006] Obtain the coordinate data of the meteorological station to be predicted, and process the coordinate data to obtain the data to be measured;
[0007] Obtain the coordinate data of each meteorological station and the actual meteorological values of each meteorological station within a historical period, preprocess the coordinate data of each meteorological station and the actual meteorological values of each meteorological station within the historical period, and then train an initial neural network based on the preprocessed coordinate data of each meteorological station and the actual meteorological values of each meteorological station within the historical period to obtain a constructed micrometeorological model;
[0008] Input the data to be measured into the micrometeorological model to obtain the predicted meteorological value of the meteorological station to be predicted.
[0009] Preferably, the processing of the coordinate data to obtain the data to be measured includes:
[0010] Converting the coordinate data of the meteorological station to be predicted into a projected coordinate system, and then performing Min-Max normalization processing to obtain the data to be measured.
[0011] Furthermore, the preprocessing of the coordinate data of each meteorological station and the actual meteorological values of each meteorological station within the historical period includes:
[0012] Converting the coordinate data of each meteorological station within the historical period into a projected coordinate system, and then performing Min-Max normalization processing;
[0013] Successively performing outlier, missing value, and data type mismatch processing on the actual meteorological values of each meteorological station.
[0014] Furthermore, training the initial neural network based on the preprocessed coordinate data of each meteorological station and the actual meteorological values of each meteorological station within the historical period to obtain a constructed micro-meteorological model, including:
[0015] Step F1: Divide the coordinate data of each meteorological station and the actual meteorological values of each meteorological station within the historical period after preprocessing into a training set and a test set;
[0016] Step F2: Construct a generator, using a convolutional neural network as the architecture of the generator. The input of the generator is the coordinate data of each meteorological station in the training set, and the output is the actual meteorological values of each meteorological station in the training set;
[0017] Step F3: Construct a discriminator, using a fully connected layer combined with a convolutional neural network as the architecture of the discriminator. The input of the discriminator is the actual meteorological values of each meteorological station in the training set and the predicted meteorological values of each meteorological station output by the generator;
[0018] Step F4: Alternately train the constructed generator and discriminator using the training set data, and then use the trained generator and discriminator as the initial micro-meteorological model;
[0019] Step F5: Use the test set to verify the initial micro-meteorological model to obtain the constructed micro-meteorological model.
[0020] The second aspect of the embodiments of the present application proposes a meteorological data prediction system based on a micro-meteorological model, including:
[0021] An acquisition module for acquiring the coordinate data of the meteorological station to be predicted and processing the coordinate data to obtain the data to be measured;
[0022] A construction module, which is used to obtain the coordinate data of each meteorological station and the actual meteorological values of each meteorological station within a historical period, preprocess the coordinate data of each meteorological station and the actual meteorological values of each meteorological station within the historical period, and then train an initial neural network based on the preprocessed coordinate data of each meteorological station and the actual meteorological values of each meteorological station within the historical period to obtain a constructed micro-meteorological model;
[0023] A prediction module, which is used to input the to-be-predicted data into the micro-meteorological model to obtain the predicted meteorological value of the to-be-predicted meteorological station.
[0024] Preferably, the acquisition module is further used for:
[0025] Convert the coordinate data of the to-be-predicted meteorological station into a projected coordinate system, and then perform Min-Max normalization processing to obtain the to-be-predicted data.
[0026] Furthermore, the construction module is further used for:
[0027] Convert the coordinate data of each meteorological station within the historical period into a projected coordinate system, and then perform Min-Max normalization processing;
[0028] Perform outlier, missing value, and data type mismatch processing on the actual meteorological values of each meteorological station in sequence.
[0029] Furthermore, the construction module is further used for:
[0030] Step E1: Divide the preprocessed coordinate data of each meteorological station and the actual meteorological values of each meteorological station within the historical period into a training set and a test set;
[0031] Step E2: Construct a generator, use a convolutional neural network as the architecture of the generator, the input of the generator is the coordinate data of each meteorological station in the training set, and the output is the actual meteorological value of each meteorological station in the training set;
[0032] Step E3: Construct a discriminator, use a fully connected layer combined with a convolutional neural network as the architecture of the discriminator, the input of the discriminator is the actual meteorological value of each meteorological station in the training set and the predicted meteorological value of each meteorological station output by the generator;
[0033] Step E4: Use the training set data to alternately train the constructed generator and discriminator, and then use the trained generator and discriminator as the initial micro-meteorological model;
[0034] Step E5: Use the test set to verify the initial micro-meteorological model to obtain the constructed micro-meteorological model.
[0035] A third aspect embodiment of the present application provides an electronic device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the method described in the embodiment of the first aspect is implemented.
[0036] A fourth aspect embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the method described in the embodiment of the first aspect is implemented.
[0037] The technical solutions provided by the embodiments of the present application at least bring the following beneficial effects:
[0038] The present application proposes a meteorological data prediction method and system based on a micro-meteorological model. The method includes: obtaining coordinate data of a meteorological station to be predicted, and processing the coordinate data to obtain data to be measured; obtaining coordinate data of each meteorological station and the actual meteorological values of each meteorological station within a historical period, and preprocessing the coordinate data of each meteorological station and the actual meteorological values of each meteorological station within the historical period. Then, based on the preprocessed coordinate data of each meteorological station and the actual meteorological values of each meteorological station within the historical period, an initial neural network is trained to obtain a constructed micro-meteorological model; the data to be measured is input into the micro-meteorological model to obtain the predicted meteorological value of the meteorological station to be predicted. The technical solution proposed by the present application realizes the construction of a micro-meteorological field model in a small-scale area, making meteorological prediction in a small-scale area more accurate and convenient.
[0039] Additional aspects and advantages of the present application will be given in part in the following description, become apparent in part from the following description, or be understood through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] The above and / or additional aspects and advantages of the present application will become apparent and be readily understood from the following description of the embodiments in conjunction with the drawings, where:
[0041] Figure 1 FIG. is a flowchart of a meteorological data prediction method based on a micro-meteorological model according to an embodiment of the present application;
[0042] Figure 2 FIG. is a structural diagram of a meteorological data prediction system based on a micro-meteorological model according to an embodiment of the present application. DETAILED DESCRIPTION
[0043] Embodiments of the present application will be described in detail below. Examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to explain the present application, and should not be construed as limiting the present application.
[0044] A meteorological data prediction method and system based on a micro-meteorological model proposed by the present application. The method includes: obtaining coordinate data of a meteorological station to be predicted, and processing the coordinate data to obtain data to be measured; obtaining coordinate data of each meteorological station and actual meteorological values of each meteorological station within a historical period, preprocessing the coordinate data of each meteorological station and the actual meteorological values of each meteorological station within the historical period, and then training an initial neural network based on the preprocessed coordinate data of each meteorological station and the actual meteorological values of each meteorological station within the historical period to obtain a constructed micro-meteorological model; inputting the data to be measured into the micro-meteorological model to obtain predicted meteorological values of the meteorological station to be predicted. The technical solution proposed by the present application realizes the construction of a micro-meteorological field model in a small-scale area, making meteorological prediction in a small-scale area more accurate and convenient.
[0045] A meteorological data prediction method and system based on a micro-meteorological model according to an embodiment of the present application will be described below with reference to the accompanying drawings.
[0046] Embodiment 1
[0047] Figure 1 FIG. is a flowchart of a meteorological data prediction method based on a micro-meteorological model according to an embodiment of the present application. As Figure 1 shown, the method includes:
[0048] Step 1: Obtain coordinate data of a meteorological station to be predicted, and process the coordinate data to obtain data to be measured;
[0049] In an embodiment of the present disclosure, the processing the coordinate data to obtain data to be measured includes:
[0050] Converting the coordinate data of the meteorological station to be predicted into a projected coordinate system, and then performing Min-Max normalization processing to obtain data to be measured.
[0051] Step 2: Obtain coordinate data of each meteorological station and actual meteorological values of each meteorological station within a historical period, preprocess the coordinate data of each meteorological station and the actual meteorological values of each meteorological station within the historical period, and then train an initial neural network based on the preprocessed coordinate data of each meteorological station and the actual meteorological values of each meteorological station within the historical period to obtain a constructed micro-meteorological model;
[0052] In the embodiments of the present disclosure, the preprocessing of the coordinate data of each meteorological station and the actual meteorological values of each meteorological station within the historical period includes:
[0053] Converting the coordinate data of each meteorological station within the historical period into a projected coordinate system, and then performing Min-Max normalization processing;
[0054] Successively performing outlier, missing value, and data type mismatch processing on the actual meteorological values of each meteorological station.
[0055] It should be noted that by sorting out different data types, meteorological data types include meteorological data such as wind direction, wind speed, temperature, humidity, air pressure, rainfall, solar radiation, etc., and the original meteorological data is classified and sorted according to different data types and data sources.
[0056] When an outlier appears, set the value to the instantaneous value of the corresponding meteorological value at the nearest moment; when a missing value appears, set the value to the instantaneous value of the corresponding meteorological value at the nearest moment; when an outlier appears, set the value to the instantaneous value of the corresponding meteorological value at the nearest moment; when data with a data type mismatch appears, convert it to a specified data format.
[0057] It should be noted that the coordinate data of the ground meteorological station is obtained and coordinate conversion is performed, and it is converted into a uniformly specified projected coordinate system, and this coordinate is used as the calculation data.
[0058] The coordinate calculation data is normalized by the Min-Max normalization method to obtain sample data.
[0059] Among them, the specific formula of the Min-Max normalization processing method is as follows:
[0060]
[0061] In the formula, x i is the characteristic parameter of the original coordinate data, x max and x min are the maximum and minimum values of the original coordinate data, and γ i is the characteristic parameter of the coordinate data after Min-Max normalization.
[0062] In the embodiments of the present disclosure, the training of the initial neural network based on the preprocessed coordinate data of each meteorological station and the actual meteorological values of each meteorological station within the historical period to obtain the constructed micro-meteorological model includes:
[0063] Step F1: Divide the coordinate data of each meteorological station and the actual meteorological values of each meteorological station within the historical period after preprocessing into a training set and a test set;
[0064] Step F2: Construct a generator, using a convolutional neural network as the architecture of the generator. The input of the generator is the coordinate data of each meteorological station in the training set, and the output is the actual meteorological value of each meteorological station in the training set;
[0065] Step F3: Construct a discriminator, using a fully connected layer combined with a convolutional neural network as the architecture of the discriminator. The input of the discriminator is the actual meteorological value of each meteorological station in the training set and the predicted meteorological value of each meteorological station output by the generator;
[0066] Step F4: Use the training set data to alternately train the constructed generator and discriminator, and then use the trained generator and discriminator as the initial micro-meteorological model;
[0067] Step F5: Use the test set to verify the initial micro-meteorological model to obtain the constructed micro-meteorological model.
[0068] It should be noted that the construction process of the micro-meteorological model specifically includes:
[0069] Step R1: Data preparation. Collect the required meteorological data and sample data of ground meteorological stations, and integrate them into a data set. In the data set, each sample should contain a set of meteorological data and the corresponding characteristic parameters of the ground meteorological station coordinates.
[0070] Step R2: Construct a generator. Set a group of neural networks, with the input being the ground meteorological station coordinates and the output being the corresponding meteorological data. The architecture of the generator adopts a convolutional neural network structure. During the training process, the goal of the generator is to generate samples similar to the training data and be as close as possible to the real meteorological data distribution.
[0071] Step R3: Construct a discriminator. The discriminator is also a group of neural networks, with the input being a set of meteorological data and the corresponding ground meteorological station coordinates, and the output being a binary classification result to judge whether the input conforms to the real meteorological data distribution. The architecture of the discriminator adopts a structure of a fully connected layer combined with a convolutional neural network. During the training process, the goal of the discriminator is to accurately distinguish between real meteorological data and meteorological data generated by the generator as much as possible.
[0072] Step R4: Train the GAN. During the training process, the generator and the discriminator are trained alternately. First, the generator takes the coordinates of the ground meteorological stations as input and generates corresponding meteorological data. Then, the discriminator takes the real meteorological data and the meteorological data generated by the generator as input and outputs the corresponding binary classification results. Next, according to the output results of the discriminator, the parameters of the generator are adjusted so that the meteorological data generated by the generator is closer to the real meteorological data distribution. At the same time, the parameters of the discriminator are adjusted so that it can more accurately distinguish between real meteorological data and the meteorological data generated by the generator.
[0073] Step R5: Evaluate the GAN. After the training is completed, a test set is used to evaluate the performance of the GAN. By inputting the coordinates of the ground meteorological stations in the test set into the generator to generate corresponding meteorological data and comparing it with the real meteorological data, the generation effect of the GAN is evaluated. The mean square error (MSE) is used as the evaluation metric to evaluate the performance of the GAN.
[0074] Step 3: Input the data to be measured into the micro-meteorological model to obtain the predicted meteorological values of the meteorological stations to be predicted.
[0075] It should be noted that the coordinate data of the point to be measured is obtained, then it is converted to the projection coordinate system and subjected to Min-Max normalization processing to obtain the data to be measured.
[0076] Taking the data to be measured as input and putting it into the high-precision micro-meteorological model, the predicted meteorological values of the data to be measured are obtained through model calculation, including meteorological data such as wind direction, wind speed, temperature, humidity, air pressure, rainfall, and solar radiation.
[0077] The meteorological data prediction method based on the micro-meteorological model proposed in this embodiment has the following advantages:
[0078] (1) The present invention realizes the construction of the micro-meteorological field model in a small-scale area, making meteorological prediction in a small-scale area more convenient.
[0079] (2) The present invention uses the mainstream GAN neural network architecture as the basis for constructing the meteorological model, and generates more realistic samples through the confrontation between the generator and the discriminator.
[0080] (3) The present invention has a high degree of automation, and the construction and prediction processes basically do not require manual intervention, and the predicted meteorological values of the target location can be directly obtained.
[0081] In summary, a meteorological data prediction method based on the micro-meteorological model proposed in this embodiment realizes the construction of the micro-meteorological field model in a small-scale area, making meteorological prediction in a small-scale area more accurate and convenient.
[0082] Embodiment 2
[0083] Figure 2 The structure diagram of a meteorological data prediction system based on a micro-meteorological model provided according to an embodiment of the present application is as follows Figure 2 shown, the system includes:
[0084] An acquisition module 100, configured to acquire coordinate data of a meteorological station to be predicted, and process the coordinate data to obtain data to be measured;
[0085] A construction module 200, configured to acquire coordinate data of each meteorological station and the actual meteorological values of each meteorological station within a historical period, preprocess the coordinate data of each meteorological station and the actual meteorological values of each meteorological station within the historical period, and then train an initial neural network based on the preprocessed coordinate data of each meteorological station and the actual meteorological values of each meteorological station within the historical period to obtain a constructed micro-meteorological model;
[0086] A prediction module 300, configured to input the data to be measured into the micro-meteorological model to obtain the predicted meteorological value of the meteorological station to be predicted.
[0087] In an embodiment of the present disclosure, the acquisition module 100 is further configured to:
[0088] Convert the coordinate data of the meteorological station to be predicted into a projected coordinate system, and then perform Min-Max normalization processing to obtain data to be measured.
[0089] In an embodiment of the present disclosure, the construction module 200 is further configured to:
[0090] Convert the coordinate data of each meteorological station within the historical period into a projected coordinate system, and then perform Min-Max normalization processing;
[0091] Perform outlier, missing value, and data type mismatch processing on the actual meteorological values of each meteorological station in sequence.
[0092] In an embodiment of the present disclosure, the construction module 200 is further configured to:
[0093] Step E1: Divide the preprocessed coordinate data of each meteorological station and the actual meteorological values of each meteorological station within the historical period into a training set and a test set;
[0094] Step E2: Construct a generator, use a convolutional neural network as the architecture of the generator, the input of the generator is the coordinate data of each meteorological station in the training set, and the output is the actual meteorological value of each meteorological station in the training set;
[0095] Step E3: Construct a discriminator. Combine a fully connected layer with a convolutional neural network as the architecture of the discriminator. The input of the discriminator is the actual meteorological values of each meteorological station in the training set and the predicted meteorological values of each meteorological station output by the generator.
[0096] Step E4: Use the training set data to alternately train the constructed generator and discriminator, and then use the trained generator and discriminator as the initial micro-meteorological model.
[0097] Step E5: Use the test set to verify the initial micro-meteorological model to obtain the constructed micro-meteorological model.
[0098] In summary, a meteorological data prediction system based on a micro-meteorological model proposed in this embodiment realizes the construction of a micro-meteorological field model in a small-scale area, making the meteorological prediction in the small-scale area more accurate and convenient.
[0099] Embodiment III
[0100] To implement the above embodiment, the present disclosure also proposes an electronic device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the method described in Embodiment I is implemented.
[0101] Embodiment IV
[0102] To implement the above embodiment, the present disclosure also proposes a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the method described in Embodiment I is implemented.
[0103] In the description of this specification, the descriptions referring to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, without conflict, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.
[0104] Any process or method description, whether in a flowchart or otherwise described herein, can be understood to represent a module, segment, or portion of code including one or more executable instructions for implementing a customized logical function or process. The scope of the preferred embodiments of the present application includes additional implementations, where functions may be performed in a substantially simultaneous manner or in a reverse order according to the functions involved, rather than in the order shown or discussed. This should be understood by those skilled in the art to which the embodiments of the present application pertain.
[0105] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present application. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.
Claims
1. A meteorological data prediction method based on a micro-meteorological model, characterized in that: The method comprises: Acquire coordinate data of the meteorological site to be predicted, and process the coordinate data to obtain data to be measured; Acquire the coordinate data of each meteorological station in the historical period and the actual meteorological value of each meteorological station, pre-process the coordinate data of each meteorological station in the historical period and the actual meteorological value of each meteorological station, and then train the initial neural network based on the pre-processed coordinate data of each meteorological station in the historical period and the actual meteorological value of each meteorological station to obtain a constructed micro-meteorological model; The data to be measured is input into the micro-meteorological model to obtain the predicted meteorological value of the meteorological site to be predicted.
2. The method according to claim 1, characterized in that The step of processing the coordinate data to obtain the data to be measured includes: The coordinate data of the meteorological station to be predicted is converted into a projection coordinate system, and then Min-Max normalization is performed to obtain the data to be measured.
3. The method according to claim 2, characterized in that The preprocessing of the coordinate data of each meteorological station in the historical period and the actual meteorological values of each meteorological station includes: The coordinate data of each meteorological station in the historical period is converted into a projected coordinate system, and then subjected to Min-Max standardization processing; The actual meteorological values of each meteorological station are processed in turn for abnormal values, missing values and data type inconsistency.
4. The method according to claim 3, characterized in that The initial neural network is trained based on the pre-processed coordinate data of each meteorological station in the historical period and the actual meteorological values of each meteorological station to obtain a constructed micro-meteorological model, including: Step F1: dividing the pre-processed coordinate data of each meteorological station in the historical period and the actual meteorological values of each meteorological station into a training set and a test set; Step F2: construct a generator, using a convolutional neural network as the architecture of the generator, wherein the generator input is the coordinate data of each meteorological station in the training set, and the output is the actual meteorological value of each meteorological station in the training set; Step F3: construct a discriminator, using a fully connected layer combined with a convolutional neural network as the architecture of the discriminator, wherein the input of the discriminator is the actual meteorological value of each meteorological station in the training set and the predicted meteorological value of each meteorological station output by the generator; Step F4: alternately train the constructed generator and discriminator using the training set data, and then use the trained generator and discriminator as the initial micrometeorological model; Step F5: Use the test set to verify the initial micro-meteorological model to obtain a constructed micro-meteorological model.
5. A meteorological data prediction system based on a micro-meteorological model, characterized in that: The system comprises: An acquisition module is used to acquire the coordinate data of the meteorological site to be predicted, and process the coordinate data to obtain the data to be measured; A construction module is used to obtain the coordinate data of each meteorological station in the historical period and the actual meteorological value of each meteorological station, and pre-process the coordinate data of each meteorological station in the historical period and the actual meteorological value of each meteorological station, and then train the initial neural network based on the pre-processed coordinate data of each meteorological station in the historical period and the actual meteorological value of each meteorological station to obtain a constructed micro-meteorological model; The prediction module is used to input the data to be measured into the micro-meteorological model to obtain the predicted meteorological value of the meteorological site to be predicted.
6. The system according to claim 5, characterized in that The acquisition module is also used for: The coordinate data of the meteorological station to be predicted is converted into a projection coordinate system, and then Min-Max normalization is performed to obtain the data to be measured.
7. The system according to claim 6, characterized in that The building blocks are also used to: The coordinate data of each meteorological station in the historical period is converted into a projected coordinate system, and then subjected to Min-Max standardization processing; The actual meteorological values of each meteorological station are processed in turn for abnormal values, missing values and data type inconsistency.
8. The system according to claim 7, characterized in that The building blocks are also used to: Step E1: dividing the pre-processed coordinate data of each meteorological station in the historical period and the actual meteorological values of each meteorological station into a training set and a test set; Step E2: construct a generator, using a convolutional neural network as the architecture of the generator, wherein the generator input is the coordinate data of each meteorological station in the training set, and the output is the actual meteorological value of each meteorological station in the training set; Step E3: construct a discriminator, using a fully connected layer combined with a convolutional neural network as the architecture of the discriminator, wherein the input of the discriminator is the actual meteorological value of each meteorological station in the training set and the predicted meteorological value of each meteorological station output by the generator; Step E4: alternately train the constructed generator and discriminator using the training set data, and then use the trained generator and discriminator as the initial micrometeorological model; Step E5: Use the test set to verify the initial micro-meteorological model to obtain a constructed micro-meteorological model.
9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method according to any one of claims 1 to 5 is implemented.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.