Middle and upper atmosphere prediction method and system, computer equipment and storage medium

Abstract

The invention relates to the technical field of weather prediction, in particular to a middle and upper atmosphere prediction method and system, computer equipment and a storage medium, and the method comprises the steps: obtaining observation data of a multi-source space environment, and calling a corresponding prediction model through a preset model scheduling strategy; based on the category of the prediction model, matching corresponding target input data in the observation data, and inputting the target input data into the prediction model to obtain a prediction result; and according to the prediction result, constructing a visual middle and upper atmosphere prediction product view, and presenting the visual middle and upper atmosphere prediction product view through a display interface of the middle and upper atmosphere prediction platform. The method has the effect of improving the accuracy of middle and upper atmosphere prediction and the display effect of the prediction result.

Description

Technical Field

[0001] The present application relates to the field of weather forecasting technology, and in particular to a method, system, computer device, and storage medium for forecasting mid- and high-level atmospheres. Background Art

[0002] Currently, mid- and upper-atmosphere forecasting is a technology that predicts various environmental conditions originating from solar activity and affecting near-Earth space and Earth's technological systems. Its main goal is to warn and quantify phenomena such as geomagnetic storms, ionospheric disturbances, and changes in upper-atmosphere density, in order to ensure the safe operation of in-orbit satellites, the accuracy and reliability of global navigation and positioning services, and the stability of transoceanic communications and power grid systems.

[0003] Existing mid- and upper-atmosphere forecasting technologies still have limitations when it comes to providing panoramic, multi-layered forecasts of the complex space environment. Most forecasting methods focus on forecasting a single physical phenomenon or a specific atmospheric region. For example, they may only provide forecasts of ionospheric disturbances or focus solely on density changes in the thermosphere. This point- or line-based forecasting model lacks an effective description of the interactions and energy coupling processes between different atmospheric layers, such as the mesosphere and thermosphere, or the thermosphere and ionosphere. Therefore, it is difficult to fully represent the entire physical chain of mid- and upper-atmospheric events, from their source to Earth's response. The forecast products provided are often independent, requiring users to independently integrate and interpret fragmented information about different physical phenomena from different systems, hindering a comprehensive understanding of the space environment. Therefore, there is room for improvement. Summary of the Invention

[0004] In order to improve the accuracy of mid- and high-level atmospheric prediction and the display effect of prediction results, the present application provides a mid- and high-level atmospheric prediction method, system, computer device and storage medium.

[0005] The above-mentioned invention objective of this application is achieved through the following technical solutions:

[0006] A mid- and high-level atmosphere prediction method is applied to a mid- and high-level atmosphere prediction platform, the mid- and high-level atmosphere prediction method comprising:

[0007] Obtain observation data from multiple sources of spatial environments and call the corresponding prediction model through the preset model scheduling strategy;

[0008] Based on the category of the prediction model, matching corresponding target input data in the observation data, inputting the target input data into the prediction model to obtain a prediction result;

[0009] Based on the prediction results, a visual mid- and high-level atmosphere prediction product view is constructed, and the visual mid- and high-level atmosphere prediction product view is presented through a display interface of the mid- and high-level atmosphere prediction platform.

[0010] By adopting the above technical solution, by acquiring observational data from multiple sources of the spatial environment and strategically invoking multiple prediction models, it is possible to systematically integrate data from different sources within a unified framework and match the most appropriate analysis tools to different prediction tasks. This overcomes the limitations of traditional methods, such as single data sources and poor model versatility, and lays the foundation for comprehensive and accurate forecasts. By matching and inputting corresponding target input data based on model categories, it ensures that each called specialized model can obtain the corresponding input data, thereby ensuring the effectiveness and accuracy of model calculations and avoiding prediction failures or reduced accuracy due to mismatched or incomplete input data. By constructing and presenting a unified visual view of mid- and upper-atmosphere forecast products, it is possible to integrate complex numerical results with different physical meanings from multiple models into an intuitive and easy-to-understand comprehensive situation map, greatly lowering the threshold for users to interpret and analyze information, significantly improving the efficiency of mid- and upper-atmosphere situational awareness and the scientific nature of decision-making.

[0011] In a preferred example, the present application may be further configured as follows: matching corresponding target input data from the observed data based on the category of the prediction model, inputting the target input data into the prediction model, and obtaining a prediction result, specifically including:

[0012] When the category of the prediction model is an ionospheric storm prediction model, the corresponding target input data matched is the first data, and the obtained prediction result is the ionospheric storm prediction result;

[0013] When the category of the prediction model is a thermospheric dynamics simulation model, the corresponding target input data matched is the third data, and the obtained prediction result is a thermospheric state prediction result;

[0014] When the category of the prediction model is an irregular body feature recognition prediction model, the corresponding target input data is matched as the second data, and the obtained prediction result is an irregular body feature recognition prediction result;

[0015] When the category of the prediction model is a mid-to-high-level atmospheric density and temperature model, the corresponding target input data matched is the fourth data, and the obtained prediction result is an atmospheric density and temperature prediction result.

[0016] By adopting the above technical solutions and dividing the prediction tasks into ionospheric storm prediction, irregular body feature identification and prediction, thermosphere state simulation and density and temperature modeling of the middle and upper atmosphere, it is possible to carry out targeted and specialized treatment of phenomena at different levels and scales in the middle and upper atmosphere, thereby achieving higher accuracy and reliability in the prediction of macro-overall disturbances, local fine risks, physical background fields and near-space environments, and realizing more refined domain and layered management and prediction of complex space environments.

[0017] In a preferred example, the present application may be further configured as follows: when the category of the prediction model is an ionospheric storm prediction model, the corresponding target input data is matched as the first data, and the obtained prediction result is the ionospheric storm prediction result, specifically including:

[0018] The first data includes at least ionospheric total electron content data, the AE index and the F10.7 solar activity index;

[0019] The ionospheric total electron content data, the AE index, and the F10.7 solar activity index are input into the trained LSTM-CNN hybrid ionospheric storm prediction model to perform time series prediction through the ionospheric storm prediction model;

[0020] The ionospheric storm prediction result is obtained from the ionospheric storm prediction model.

[0021] By adopting the above technical solution and using the LSTM-CNN hybrid model to perform time series prediction on spatiotemporal data including TEC, AE and F10.7, it is possible to simultaneously use the CNN network to extract the spatial morphological features in the TEC map and the LSTM network to capture the temporal evolution laws and long-range dependencies of various physical quantities. This can provide a deeper understanding of the complete physical process of ionospheric storms than a single network structure, significantly improving the prediction accuracy of the entire process of storm occurrence, development and extinction.

[0022] In a preferred example, the present application may be further configured as follows: when the category of the prediction model is an irregular body feature recognition prediction model, the corresponding target input data is matched as the second data, and the obtained prediction result is an irregular body feature recognition prediction result, specifically including:

[0023] The second data includes at least GNSS observation data, optical observation data, ionospheric detection parameters and radar detection data;

[0024] Based on the pre-acquired edge extraction algorithm, phase coherence analysis algorithm, and machine learning algorithm, the irregular body feature recognition and prediction model is constructed, and the GNSS observation data, the optical observation data, the ionospheric sounding parameters, and the radar detection data are input into the irregular body feature recognition and prediction model to perform irregular body recognition and short-, medium-, and long-term forecasts through the irregular body feature recognition and prediction model;

[0025] The irregular body feature recognition and prediction result is obtained from the irregular body feature recognition and prediction model, and the irregular body feature recognition and prediction result at least includes an irregular body feature recognition result and an irregular body prediction result.

[0026] By adopting the above technical solution, a prediction model constructed by combining edge extraction algorithm, phase coherence analysis algorithm and machine learning algorithm is used to identify irregular bodies in multi-source heterogeneous diagnostic data including GNSS, optical, radar, etc., which can quickly mine complex nonlinear patterns related to ionospheric irregularities from massive, high-dimensional features with extremely high computational efficiency, thereby achieving rapid and accurate probabilistic forecasts of local and short-term signal scintillation risks, and providing tactical-level avoidance guidance for applications such as drones and high-precision positioning that have strict requirements on signal quality.

[0027] In a preferred example, the present application may be further configured as follows: when the category of the prediction model is a thermospheric dynamics simulation model, the corresponding target input data is matched as the third data, and the obtained prediction result is a thermospheric state prediction result, specifically including:

[0028] The third data at least includes solar activity parameters, geomagnetic disturbance parameters and time parameters;

[0029] The solar activity parameter, the geomagnetic disturbance parameter and the time parameter are input into the thermospheric dynamics simulation model based on TIEGCM, so as to predict the thermospheric electric field and electric potential through the thermospheric dynamics simulation model;

[0030] The thermospheric state prediction result is obtained from the thermospheric dynamics simulation model.

[0031] By adopting the above technical solution, by using a physical model based on TIEGCM and inputting external driving parameters such as solar activity and geomagnetic disturbances to predict the thermospheric electric field and potential, it is possible to simulate the global electrodynamic process driven by external energy by solving the first-principles physical equations while ensuring the self-consistency of the physical mechanism, thereby providing a key, upstream physical driving field for understanding and predicting the transport and distribution of ionospheric plasma, and improving the physical completeness of the entire prediction system.

[0032] In a preferred example, the present application may be further configured as follows: when the category of the prediction model is a mid-to-high-level atmospheric density and temperature model, the corresponding target input data is matched as the fourth data, and the obtained prediction result is an atmospheric density and temperature prediction result, specifically including:

[0033] The fourth data at least includes solar activity parameters, geomagnetic activity index, occultation atmosphere temperature and occultation atmosphere density;

[0034] The solar activity parameters, geomagnetic activity index, occultation atmosphere temperature and occultation atmosphere density are input into the trained Light-GBM mid- and high-level atmospheric density and temperature model to predict the near-space environment.

[0035] The atmospheric density and temperature prediction results are obtained from the mid- and high-level atmospheric density and temperature model.

[0036] By adopting the above technical solution, by using the density and temperature model of the middle and upper atmosphere, and integrating background driving parameters such as the sun and geomagnetism with real-time observation data such as occultation atmospheric temperature and density, the global stability of the empirical model or background model can be combined with the accuracy of real-time high-precision observation data to achieve complementary advantages, thereby generating a more reliable and accurate near-space environment forecast than a single type of model, providing more valuable decision support for applications such as hypersonic aircraft.

[0037] In a preferred example, the present application may be further configured as follows: constructing a visual mid- and upper-atmosphere forecast product view based on the forecast results, specifically including:

[0038] Organizing the ionospheric storm prediction results, the irregular body disturbance prediction results, the thermosphere state prediction results, and the atmospheric density and temperature prediction results according to spatial regions and time scales to obtain organized prediction layer data;

[0039] Mapping the predicted layer data to preset product layers, wherein the product layers include an ionosphere layer, an irregular body disturbance layer, a thermosphere layer, and a mid- and high-level atmosphere layer;

[0040] Based on preset layer display rules, the product layer is subjected to layer synthesis processing to generate the visualized mid- and upper-level atmosphere forecast product view.

[0041] By adopting the above technical solution and partitioning the prediction results according to the spatiotemporal scale, massive continuous data fields can be preprocessed into discrete data units that are easy to quickly index and retrieve, thereby greatly improving the system's data query performance when responding to user interaction operations, laying the foundation for a smooth visualization experience. By mapping data to preset product layers and performing layer synthesis processing based on preset layer display rules, a visual view of the mid- and high-level atmospheric forecast product is obtained, which elevates the system from a static data display tool to a powerful platform that can support active and exploratory data analysis.

[0042] The second object of the present invention is achieved through the following technical solutions:

[0043] A mid- and high-level atmospheric prediction system, comprising:

[0044] The data acquisition management module is used to obtain observation data from multiple sources of spatial environments and call the corresponding prediction model through the preset model scheduling strategy;

[0045] A prediction result generation module is used to match corresponding target input data in the observation data based on the category of the prediction model, input the target input data into the prediction model, and obtain a prediction result;

[0046] The visualization and presentation module is used to construct a visualized mid- and high-level atmosphere prediction product view based on the prediction results, and present the visualized mid- and high-level atmosphere prediction product view through the display interface of the mid- and high-level atmosphere prediction platform.

[0047] By adopting the above technical solution, by acquiring observational data from multiple sources of the spatial environment and strategically invoking multiple prediction models, it is possible to systematically integrate data from different sources within a unified framework and match the most appropriate analysis tools to different prediction tasks. This overcomes the limitations of traditional methods, such as single data sources and poor model versatility, and lays the foundation for comprehensive and accurate forecasts. By matching and inputting corresponding target input data based on model categories, it ensures that each called specialized model can obtain the corresponding input data, thereby ensuring the effectiveness and accuracy of model calculations and avoiding prediction failures or reduced accuracy due to mismatched or incomplete input data. By constructing and presenting a unified visual view of mid- and upper-atmosphere forecast products, it is possible to integrate complex numerical results with different physical meanings from multiple models into an intuitive and easy-to-understand comprehensive situation map, greatly lowering the threshold for users to interpret and analyze information, significantly improving the efficiency of mid- and upper-atmosphere situational awareness and the scientific nature of decision-making.

[0048] The third objective of this application is achieved through the following technical solutions:

[0049] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned method for predicting the middle and upper atmosphere are implemented.

[0050] The fourth objective of this application is achieved through the following technical solutions:

[0051] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned method for predicting the middle and upper atmosphere.

[0052] In summary, this application includes at least one of the following beneficial technical effects:

[0053] 1. By acquiring multi-source spatial environmental observation data and strategically invoking multiple prediction models, it is possible to systematically integrate data from different sources within a unified framework and match the most appropriate analysis tools to different prediction tasks. This overcomes the limitations of traditional methods, such as single data sources and poor model versatility, and lays the foundation for comprehensive and accurate forecasts. By matching and inputting corresponding target input data based on model categories, it ensures that each invoked specialized model receives the corresponding input data, thus ensuring the effectiveness and accuracy of model calculations and avoiding prediction failures or reduced accuracy due to mismatched or incomplete input data. By constructing and presenting a unified visual view of mid- and upper-atmosphere forecast products, it can integrate complex numerical results with different physical meanings from multiple models into an intuitive and easy-to-understand comprehensive situation map, greatly reducing the threshold for users to interpret and analyze information, significantly improving the efficiency of mid- and upper-atmosphere situation awareness and the scientific nature of decision-making.

[0054] 2. By partitioning and organizing the forecast results according to the spatiotemporal scale, massive continuous data fields can be preprocessed into discrete data units that are easy to quickly index and retrieve, thereby greatly improving the system's data query performance when responding to user interactions and laying the foundation for a smooth visualization experience. By mapping data to preset product layers and performing layer synthesis processing based on preset layer display rules, a visual view of the mid- and upper-level atmospheric forecast products is obtained, which elevates the system from a static data display tool to a powerful platform that can support active and exploratory data analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a flowchart of an implementation method of a mid- and high-level atmosphere prediction method in one embodiment of the present application;

[0056] Figure 2This is a flowchart for implementing step S20 in a method for predicting the middle and upper atmosphere in one embodiment of the present application;

[0057] Figure 3 This is a flowchart for implementing step S21 in a method for predicting the middle and upper atmosphere in one embodiment of the present application;

[0058] Figure 4 This is a flowchart for implementing step S22 in a method for predicting the middle and upper atmosphere in one embodiment of the present application;

[0059] Figure 5 This is a flowchart for implementing step S23 in a method for predicting the middle and upper atmosphere in one embodiment of the present application;

[0060] Figure 6 This is a flowchart for implementing step S24 in a method for predicting the middle and upper atmosphere in one embodiment of the present application;

[0061] Figure 7 This is a flowchart for implementing step S30 in a method for predicting the middle and upper atmosphere in one embodiment of the present application;

[0062] Figure 8 This is a principle block diagram of a mid- and high-level atmosphere prediction system in one embodiment of the present application;

[0063] Figure 9 It is a schematic diagram of the internal structure of a computer device in one embodiment of the present application. DETAILED DESCRIPTION

[0064] The following examples will help those skilled in the art further understand the purpose of this application, but are not intended to limit this application in any form. It should be noted that those skilled in the art may make several modifications and improvements without departing from the scope of this application. These modifications and improvements are all within the scope of this application.

[0065] In the following description, specific details such as specific system structures and techniques are provided for purposes of illustration rather than limitation to facilitate a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, systems, circuits, and methods are omitted to avoid obscuring the description of the present application with unnecessary detail.

[0066] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.

[0067] The present application is further described in detail below with reference to the accompanying drawings.

[0068] In one embodiment, if Figure 1 As shown, the present application discloses a mid- and high-level atmospheric prediction method, which is applied to a mid- and high-level atmospheric prediction platform and specifically includes the following steps:

[0069] S10: Obtain observation data of the multi-source spatial environment and call the corresponding prediction model through the preset model scheduling strategy.

[0070] Specifically, obtaining multi-source space environment observation data refers to retrieving timely and representative data content from multiple space environment detection platforms or data centers, including but not limited to ionospheric TEC data obtained through ground-based GNSS station networks, ionospheric state data collected by electron density monitors on satellite orbits, geomagnetic disturbance parameters such as Kp, Ap, and Sym-H index collected in real time by geomagnetic arrays, solar activity data such as F10.7 solar radio flux, sunspot number, solar proton event bulletin, EUV / FUV flux obtained through public services such as NOAA, NASA, or SEPC, and atmospheric temperature, density, wind speed, water vapor, ozone, OH, and other middle and upper atmospheric physical quantities measured by satellites such as TIMED, COSMIC, and FY. In actual operation, these observation data are first jointly normalized by timestamp and spatial coordinates, and classified according to the physical type, measurement space range, and time window of the data, so that subsequent models can schedule and call the required data subset on demand to carry out middle and upper atmospheric prediction analysis.

[0071] S20: Based on the category of the prediction model, the corresponding target input data is matched in the observation data, and the target input data is input into the prediction model to obtain the prediction result.

[0072] Specifically, the calling method based on the model scheduling strategy refers to obtaining the input data required by the prediction model according to the category of the existing prediction model under the guidance of a set of data-model mapping relationship tables or scheduling rules built into the platform. Assuming that the model scheduling strategy triggers the calling of the ionospheric storm prediction model and the irregular body feature recognition and prediction model at the same time, then for the ionospheric storm prediction model, it will accurately match and extract the time series data of multiple variables such as F10.7 index, Kp index, global TEC grid, etc. from the massive observation data according to its category characteristics, and construct them into a time series prediction model that meets the requirements of the prediction model. The tensor with time dimension required by the type input; and for the irregular body feature recognition and prediction model, a set of completely different feature data focusing on local diagnosis, such as S4 scintillation index, airglow image features, radar echo power spectrum, etc., will be matched and extracted according to its category characteristics, and they will be constructed into a feature vector suitable for classification model and without time dimension. Through this on-demand matching mechanism based on model category, highly differentiated and specialized data utilization of the same comprehensive data pool is realized, and the prepared target input data is sent to the corresponding model to drive it to complete a prediction and output the result.

[0073] S30: Based on the prediction results, a visual mid- and high-level atmosphere prediction product view is constructed, and the visual mid- and high-level atmosphere prediction product view is presented through a display interface of the mid- and high-level atmosphere prediction platform.

[0074] Specifically, the prediction results output by each model in the previous step are processed. First, all gridded prediction results based on geographic coordinates, such as the 2D TEC matrix and the 3D thermospheric density field, are cut according to preset geographic tile specifications such as 2°×2° longitude and latitude and a unified time step such as 1 hour, and encapsulated into a structured data object containing a geographic location index, a time index, and a prediction value matrix. Then, these structured data objects are classified according to their physical properties such as ionospheric TEC, thermospheric density, and irregular body probability and mapped to different product layers. For example, all structured data objects for TEC predictions are organized into the ionospheric layer, and a set of color mapping tables from blue to red are pre-defined for it to represent the high and low TEC values. Ultimately, based on user interactions such as dragging the map, scrolling the time slider, or system-preset display rules such as a rendering rule that includes layer stacking order, transparency, and color blending mode, by overlaying different layers in a meaningful way, such as overlaying the highlighted risk area layer on the physical background parameter layer, a visual mid- and high-level atmosphere forecast product view with rich content, clear logic, and user-interactive operations such as zooming, panning, and time relaxation is generated, and the view is pushed to the display interface of the mid- and high-level atmosphere forecast platform through real-time communication protocols such as WebSocket.

[0075] In one embodiment, if Figure 2 As shown, in step S20, based on the category of the prediction model, the corresponding target input data is matched in the observation data, and the target input data is input into the prediction model to obtain the prediction result, which specifically includes:

[0076] S21: When the category of the prediction model is an ionospheric storm prediction model, the corresponding target input data is matched as the first data, and the obtained prediction result is the ionospheric storm prediction result.

[0077] Specifically, when the scheduled model category is identified as an ionospheric storm prediction model, the model is a deep learning network based on the bidirectional long short-term memory network (Bi-directionalLSTM), which performs a targeted data matching and combination action to obtain the first data from the observation data. The first data is specifically a multivariate time series data set, which integrates the F10.7 solar activity index and Kp, AE geomagnetic disturbance index as external drivers in the time dimension, as well as the global TEC grid data and satellite occultation electron density profile as representations of the current state of the ionosphere. After the first data is input into the model, the ionospheric storm prediction results generated by the model are a set of comprehensive products that include TEC and three-dimensional electron density evolution forecasts for the next few days, and are accompanied by key event markers such as the start and peak time of the storm.

[0078] S22: When the category of the prediction model is an irregular body feature recognition prediction model, the corresponding target input data is matched as the second data, and the obtained prediction result is an irregular body feature recognition prediction result.

[0079] Specifically, when the scheduled model category is identified as an irregular body feature recognition and prediction model, a data matching action is performed for multiple diagnostic features to obtain second data from the observation data. The second data specifically includes GNSS carrier phase data and S4 scintillation index from multiple sources such as WHU, CETC, CMPII and COSMIC, airglow image features processed by the CMPII all-sky imager, the F2 layer critical frequency and corresponding altitude from the GIRO global ionospheric altimeter observation network, and plasma parameters and echo characteristics detected by coherent / incoherent scattering radars from institutions such as IGGCAS. By inputting the second data into the trained irregular body feature recognition and prediction model, and utilizing its advantages of being able to process massive amounts of data at high speed and perform high-precision classification, a probability value representing the occurrence of severe disturbances in the geographic grid in the future is quickly output, and the disturbance drift vector field is given in combination with the radar velocity measurement data, which together constitute the irregular body disturbance prediction result.

[0080] S23: When the category of the prediction model is a thermospheric dynamics simulation model, the corresponding target input data is matched as the third data, and the obtained prediction result is a thermospheric state prediction result.

[0081] Specifically, when the scheduled model category is identified as a thermospheric dynamics simulation model, a full-scale data matching action required to drive a complex physical model is performed to obtain third data from the observation data. The third data specifically includes geomagnetic and solar activity indices such as F10.7, Ap, and Dst. During the simulation process, a large amount of in-situ observation data measured at different altitudes such as 507km and 800-850km from multiple in-orbit satellites such as NOAA and CSES will be continuously integrated. These in-situ observation data specifically cover physical parameters such as in-situ particles, magnetic fields, electric fields, electric potentials, electrical conductivity, settling particles, FUV / EUV radiation, Joule heating, neutral component density, and ion flux. After the third data is input into the model, a global and three-dimensional dynamic thermospheric state prediction result is obtained.

[0082] S24: When the category of the prediction model is a mid-to-high-level atmospheric density and temperature model, the corresponding target input data is matched as the fourth data, and the obtained prediction result is an atmospheric density and temperature prediction result.

[0083] Specifically, when the model category being scheduled is a mid- and high-level atmospheric density and temperature model for refined forecasting of the near-space environment, a composite data matching action driving the hybrid forecasting model is executed to obtain fourth data from the observation data. The fourth data specifically includes 20-80km temperature, water vapor, and ozone profile data from multiple sources such as TIMED / SABER, CMA, and NOAA, 20-80km wind field data, and occultation bending angle or refractive index data from multiple sources such as COSMIC and CMA. After the fourth data is input into the model, the atmospheric density and temperature prediction results obtained are a set of probabilistic forecast products containing uncertainty information, which not only provides the most likely estimates of future near-space environments such as temperature, density, and ozone concentrations, but also provides the credibility of the forecast in the form of ensemble discreteness.

[0084] In one embodiment, if Figure 3 As shown, in step S21, that is, when the category of the prediction model is the ionospheric storm prediction model, the corresponding target input data is matched as the first data, and the obtained prediction result is the ionospheric storm prediction result, which specifically includes:

[0085] S211: The first data includes at least ionospheric total electron content data, AE index and F10.7 solar activity index.

[0086] Specifically, the first data may be the total electron content data of the ionosphere, the AE index and the F10.7 solar activity index. The F10.7 solar activity index is an effective proxy indicator of solar extreme ultraviolet radiation. Its function is to define the basic state or background density of the ionosphere. The AE index is the auroral electrocurrent index. Its function is to characterize the direct driving force of energy injection in high-latitude regions. The total electron content data of the ionosphere provides real-time observations of the real-time response of the ionosphere to the above-mentioned driving forces. It intuitively shows where the ionosphere becomes thicker, i.e., positive phase bursts, or thinner, i.e., negative phase bursts.

[0087] S212: Through the trained LSTM-CNN hybrid ionospheric storm prediction model, the ionospheric total electron content data, AE index and F10.7 solar activity index are input into the ionospheric storm prediction model to perform time series prediction through the ionospheric storm prediction model.

[0088] Specifically, the LSTM-CNN hybrid ionospheric storm prediction model is a prediction unit that combines the advantages of two different neural networks and has been trained on a large amount of historical data. It can simultaneously and efficiently process the spatial and temporal features contained in the input data, thereby achieving better prediction performance than a single network structure. First, the ionospheric total electron content data, which is input at each time step and present as a two-dimensional image or grid, is processed by the convolutional neural network (CNN) component of the model. Leveraging the CNN's powerful spatial feature extraction capabilities, it automatically learns and identifies key spatial morphological features in the TEC map, such as the positive phase burst region and the ionospheric trough, and compresses this complex spatial information into a low-dimensional feature vector. Next, the series of feature vectors extracted by the CNN at multiple consecutive time steps are constructed into a time series and input into the long short-term memory (LSTM) component of the model. Leveraging the LSTM's ability to remember and understand long-range temporal dependencies, it learns and predicts the dynamic evolution of these spatial features. Through this collaborative working mode, with the CNN responsible for understanding space and the LSTM responsible for understanding time, a high-quality time series forecast is ultimately achieved.

[0089] S213: Obtain ionospheric storm prediction results from the ionospheric storm prediction model.

[0090] Specifically, the original, high-dimensional numerical matrix output by the model is converted into an application-oriented product that is more user-friendly and has higher information density. The obtained prediction results can specifically include the following parts: one is the quantitative physical quantity forecast, such as the hourly evolution map of the global total electron content (TEC) in the next 120 hours, the three-dimensional global electron density grid data, and the F2 layer critical frequency (foF2) distribution map converted from the peak electron density; the other is the event-type forecast generated by algorithm post-processing, such as automatically identifying and marking the expected ionospheric storm start time, peak moment, expected intensity level such as moderate, strong, and storm phase type such as positive phase storm or negative phase storm and other key warning information by analyzing the shape of the TEC forecast curve, thereby providing intuitive and clear decision-making basis for applications such as satellite navigation and shortwave communication.

[0091] In one embodiment, if Figure 4 As shown, in step S22, that is, when the category of the prediction model is the irregular body feature recognition prediction model, the corresponding target input data is matched as the second data, and the obtained prediction result is the irregular body feature recognition prediction result, which specifically includes:

[0092] S221: The second data includes at least GNSS observation data, optical observation data, ionospheric detection parameters and radar detection data.

[0093] Specifically, GNSS observation data refers to carrier phase and S4 indicators, optical observation data refers to the brightness distribution of the all-sky glow imager, ionospheric detection parameters refer to the critical frequency and peak height of the F2 layer, and radar detection data refers to the coherent or incoherent scattering power spectrum characteristics. This combination can comprehensively reflect the intensity and dynamics of the disturbance body in space.

[0094] S222: Based on the pre-acquired edge extraction algorithm, phase coherence analysis algorithm and machine learning algorithm, an irregular body feature recognition and prediction model is constructed, and GNSS observation data, optical observation data, ionospheric detection parameters and radar detection data are input into the irregular body feature recognition and prediction model to identify irregular bodies and make short-, medium- and long-term forecasts through the irregular body feature recognition and prediction model.

[0095] Specifically, the irregular body feature recognition and prediction model does not rely on a single algorithm. Instead, it constructs a multi-stage composite processing pipeline that integrates the advantages of multiple algorithms. It first uses traditional algorithms with clear physical meaning to enhance the features of raw data, extracting key, interpretable intermediate features. It then leverages the powerful nonlinear fitting capabilities of machine learning algorithms to perform deep pattern recognition and prediction on these enhanced features. First, edge extraction algorithms, such as the Canny operator, are applied to optical observation data, namely airglow images, to automatically identify and quantify the morphological characteristics of dark streaks in plasma bubbles. Simultaneously, phase coherence analysis algorithms are applied to GNSS observation data to accurately quantify the severity of signal phase jitter. These features extracted by the physical algorithm, along with other directly observed parameters such as F2 layer height and radar echo power, are fed as input into a machine learning algorithm, such as a gradient boosting decision tree or a deep neural network, for final recognition and prediction.

[0096] S223: Obtaining irregular body feature recognition and prediction results from the irregular body feature recognition and prediction model, wherein the irregular body feature recognition and prediction results at least include irregular body feature recognition results and irregular body prediction results.

[0097] Specifically, irregular bodies are identified and short-, medium- and long-term forecasts are made through the irregular body feature recognition and forecasting model, obtaining a rich irregular body feature recognition and forecasting result covering both current status recognition and future forecasts, providing users with a complete information closed loop from current situation awareness to future risk prediction. The irregular body feature recognition result can be an analysis diagram that highlights the spatial range of existing irregular bodies such as plasma bubbles, and is accompanied by a parameter list that contains a quantitative description of the irregular body, such as its maximum intensity, coverage area, and internal structure complexity. The irregular body forecast result can be a future-oriented prediction product, for example, a probability distribution map showing the occurrence of significant signal flashes in various regions within the next 0-3 hours, and a dynamic drift vector field that predicts how the identified irregular bodies will move in the future.

[0098] In one embodiment, if Figure 5 As shown, in step S23, that is, when the category of the prediction model is a thermospheric dynamics simulation model, the corresponding target input data is matched as the third data, and the obtained prediction result is a thermospheric state prediction result, which specifically includes:

[0099] S231: The third data at least includes solar activity parameters, geomagnetic disturbance parameters and time parameters.

[0100] Specifically, the solar activity parameters in the third data, such as the F10.7 index or EUV spectral flux, are used to set the basic energy source of the entire simulation, which determines the background temperature and degree of ionization of the upper atmosphere. The geomagnetic perturbation parameters, such as the Kp index or the Ap index, are used to set the disturbance energy source during the simulation. They quantify the energy injection intensity from the magnetosphere and are the key to driving violent phenomena such as thermospheric storms. The time parameters including year, month, day, and universal time are used to set the time and space benchmark of the simulation, which determines the Earth's sunshine conditions under solar radiation, seasonal effects, and the duration of disturbance events. These three together constitute the necessary conditions for driving the reasonable evolution of the physical model.

[0101] S232: Obtain a thermospheric dynamics simulation model based on TIEGCM, input solar activity parameters, geomagnetic disturbance parameters, and time parameters into the thermospheric dynamics simulation model, and predict the thermospheric electric field and electric potential through the thermospheric dynamics simulation model.

[0102] Specifically, the TIEGCM-based thermospheric dynamics simulation model is a fully validated and parameter-optimized complex physical model unit that performs calculations by solving a system of partial differential equations describing the fluid mechanics, thermodynamics, and electrodynamics of the upper atmosphere. By directly simulating physical processes, it can generate a three-dimensional atmospheric state that is continuous in time and space, with mutually coupled physical quantities and completely self-consistent internal logic. After receiving the input parameters, the model calculates the global circulation, or wind field, of neutral atmospheric elements such as oxygen atoms and nitrogen molecules under the influence of solar radiation heating and collisions with ions, or ion drag, through numerical integration on a global three-dimensional grid. At the same time, the model also calculates the thermospheric electric field and corresponding electric potential distribution generated by the neutral wind field through the atmospheric wind generator effect when it cuts through the geomagnetic field. This predicted electric field and potential are key physical quantities that determine the subsequent transport and distribution of ionospheric plasma, thus constituting an important upstream link in the forecast of the entire middle and upper atmospheric system.

[0103] S233: Obtain thermospheric state prediction results from the thermospheric dynamics simulation model.

[0104] Specifically, the thermospheric state prediction result is a high-resolution and high-fidelity digital twin of the upper atmosphere. The obtained results may specifically include the following parts: first, the three-dimensional thermospheric atmospheric density field, which is the core input with the highest accuracy requirements for atmospheric drag force calculation and orbit prediction of low-orbit spacecraft; second, the three-dimensional thermospheric temperature field and wind field, including latitudinal, longitudinal and vertical wind speeds; third, key energy balance parameters, such as the Joule heating power distribution diagram, which can be used to diagnose the energy deposition process during geomagnetic storms; fourth, application-oriented derivative products, such as the total density disturbance level map simplified to low, medium and high levels at specific altitudes of 250km, 400km, etc., to facilitate users to quickly assess risks.

[0105] In one embodiment, if Figure 6 As shown, in step S24, that is, when the category of the prediction model is the middle and upper atmospheric density temperature model, the corresponding target input data is matched as the fourth data, and the obtained prediction result is the atmospheric density temperature prediction result, which specifically includes:

[0106] S241: The fourth data at least includes solar activity parameters, geomagnetic activity index, occultation atmosphere temperature and occultation atmosphere density.

[0107] Specifically, solar activity parameters and geomagnetic activity indices are mainly used as background driving parameters to drive the empirical or climatological part of the model to quickly generate a globally covered, stable benchmark atmospheric state, while occultation atmospheric temperature and occultation atmospheric density are used as real-time constraint data. They are derived from GNSS occultation detection technology with high vertical resolution and high precision, and can provide ground-truth information about the current atmospheric thermodynamic structure, which is used to correct and pull back theoretical results generated purely by background driving, ensuring that the prediction can reflect real weather changes.

[0108] S242: Obtain the trained Light-GBM mid- and high-level atmospheric density and temperature model, and input solar activity parameters, geomagnetic activity index, occultation atmospheric temperature, and occultation atmospheric density into the mid- and high-level atmospheric density and temperature model to predict the near-space environment using the mid- and high-level atmospheric density and temperature model.

[0109] Specifically, the mid- and upper-level atmospheric density and temperature model is a composite prediction unit that may contain multiple computational components and has been trained or tuned with extensive data. This model leverages the computational efficiency and global stability of empirical models with the high precision and dynamics of real-time observational data, generating more reliable predictions than single-type models. Upon receiving input data, an empirical model component within the model first rapidly calculates a global background field using solar activity parameters and geomagnetic activity indices. Simultaneously, another data-driven correction component within the model, such as a pre-trained neural network, compares high-resolution occultation atmospheric temperature and density data with the background field at corresponding points in time and space, calculating the deviation between the two. Finally, this deviation, calculated based on real observations, is superimposed on the background field in the form of a correction field, resulting in a final near-space environment prediction—the atmospheric density and temperature prediction—that is both globally accurate and highly consistent with real observations in local detail.

[0110] S243: Obtain atmospheric density and temperature prediction results from the middle and upper atmospheric density and temperature model.

[0111] Specifically, the atmospheric density and temperature prediction results are a set of probabilistic forecast products that provide users with the risk assessment information needed for decision-making. The obtained results can specifically include the following parts: first, the ensemble mean forecast field, that is, the average of the results of all ensemble members as an estimate of future atmospheric conditions such as temperature, density, air pressure, wind speed, and major components such as OH and O atoms; second, the ensemble dispersion or standard deviation field, that is, the degree of dispersion of the results of all ensemble members. The larger the value, the higher the uncertainty of the forecast and the lower the credibility; third, the probability map of specific events. For example, the percentage of members whose density exceeds a certain threshold among all ensemble members is calculated, thereby obtaining a probability map of the atmospheric density exceeding the dangerous threshold at a certain time and place in the future.

[0112] In one embodiment, if Figure 7 As shown, in step S30, based on the prediction results, a visual mid- and upper-atmosphere prediction product view is constructed, specifically including:

[0113] S31: The ionospheric storm prediction results, irregular body feature prediction results, thermosphere state prediction results and atmospheric density and temperature prediction results are partitioned and sorted according to spatial regions and time scales to obtain sorted prediction layer data.

[0114] Specifically, the prediction results output by each model in the previous step are processed. First, the prediction results are preprocessed into a series of discretized, miniaturized structured data objects with standardized spatiotemporal indexes, namely prediction layer data. For example, a three-dimensional thermal layer density prediction result covering the entire world with a forecast period of 72 hours. If the entire file is read and cut for each user request, it will cause a delay of several seconds or even longer. However, by pre-partitioning and organizing according to spatial regions and time scales, it is preprocessed into tens of thousands of independent small files or cache objects. Each object precisely corresponds to a specific geographic grid such as 2°×2° and a specific time point such as 1 hour. When a user requests to view the situation in a local area, the background can directly locate and read dozens of corresponding data objects in parallel through the spatiotemporal index, thereby reducing the delay of data retrieval to milliseconds.

[0115] S32: Mapping the prediction layer data to the preset product layers, the product layers include the ionosphere layer, the irregular body disturbance layer, the thermosphere layer and the middle and upper atmosphere layer.

[0116] Specifically, the product layer is a logical container for business applications associated with specific physical quantities or forecast products. It decouples the underlying physical storage of data from the upper-level business logic calls, making the entire system highly modular and scalable. All prediction layer data generated in the above steps that contain total electron content (TEC) forecast values, regardless of which specific model they come from, will be logically classified and mapped to the ionosphere layer. All prediction layer data that contain scintillation probability will be mapped to the irregular body disturbance layer. At the same time, when defining each product layer, some default rendering properties will be pre-bound to it. For example, a color mapping table from cold to warm tones is preset for the thermosphere layer to intuitively display the change in density from low to high.

[0117] S33: Based on the preset layer display rules, the product layers are synthesized to generate a visual mid- and upper-level atmospheric forecast product view.

[0118] Specifically, the layer display rules predetermine the spatial range, time window, layer priority, transparency and overlay mode. The application first reads the longitude and latitude boundaries of the user view and the selected prediction time as dynamic parameters, and then retrieves the ionosphere layer, irregular body disturbance layer, thermosphere layer and middle and upper atmosphere layer that match the spatial range and time window from the layer warehouse. Each layer is sent to the WebGL rendering pipeline according to priority from low to high, and alpha blending is performed according to transparency, and color fusion is controlled according to the overlay mode. After rendering is completed, a single bitmap containing the overlay effect of multiple physical quantities is output and pushed to the front-end canvas. At the same time, the time slider and layer switch control are loaded for user interaction, thereby generating a visual view of the middle and upper atmosphere forecast product.

[0119] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0120] In one embodiment, a mid- and high-level atmosphere prediction system is provided, which corresponds one-to-one to the mid- and high-level atmosphere prediction method in the above embodiment. Figure 8 As shown in Figure 1, the mid- and upper-atmosphere forecasting system includes a data acquisition and management module, a forecast result generation module, and a visualization and presentation module. The functional modules are described in detail below:

[0121] The data acquisition management module is used to obtain observation data from multiple sources of spatial environments and call the corresponding prediction model through the preset model scheduling strategy;

[0122] The prediction result generation module is used to match the corresponding target input data in the observation data based on the category of the prediction model, input the target input data into the prediction model, and obtain the prediction result;

[0123] The visualization and presentation module is used to construct a visual view of the mid- and upper-level atmosphere prediction product based on the prediction results, and present the visual view of the mid- and upper-level atmosphere prediction product through the display interface of the mid- and upper-level atmosphere prediction platform.

[0124] Optional prediction result generation module, specifically including:

[0125] The ionospheric storm prediction submodule is used to match the corresponding target input data as the first data when the category of the prediction model is the ionospheric storm prediction model, and the obtained prediction result is the ionospheric storm prediction result;

[0126] The irregular body recognition submodule is used to match the corresponding target input data as the second data when the category of the prediction model is the irregular body feature recognition prediction model, and the obtained prediction result is the irregular body feature recognition prediction result;

[0127] The thermosphere state simulation submodule is used to match the corresponding target input data as the third data when the category of the prediction model is the thermosphere dynamics simulation model, and the obtained prediction result is the thermosphere state prediction result;

[0128] The middle atmosphere modeling submodule is used to match the corresponding target input data as the fourth data when the category of the prediction model is the middle and upper atmospheric density and temperature model, and the obtained prediction result is the atmospheric density and temperature prediction result.

[0129] Optional, ionospheric storm prediction submodule, specifically including:

[0130] an ionospheric state data unit, for which the first data includes at least ionospheric total electron content data, AE index, and F10.7 solar activity index;

[0131] The model prediction unit is used to input the ionospheric total electron content data, AE index and F10.7 solar activity index into the ionospheric storm prediction model through the trained LSTM-CNN hybrid ionospheric storm prediction model, so as to perform time series prediction through the ionospheric storm prediction model;

[0132] The prediction result acquisition unit is used to obtain the ionospheric storm prediction result from the ionospheric storm prediction model.

[0133] Optional, irregular body recognition submodule, specifically including:

[0134] Irregular volume data unit, used for second data including at least GNSS observation data, optical observation data, ionospheric detection parameters and radar detection data;

[0135] A model prediction unit is used to construct an irregular body feature recognition and prediction model based on a pre-acquired edge extraction algorithm, a phase coherence analysis algorithm, and a machine learning algorithm, and input GNSS observation data, optical observation data, ionospheric sounding parameters, and radar sounding data into the irregular body feature recognition and prediction model to identify irregular bodies and perform short-, medium-, and long-term forecasts through the irregular body feature recognition and prediction model;

[0136] The prediction result acquisition unit is used to acquire the irregular body feature recognition prediction result from the irregular body feature recognition prediction model, wherein the irregular body feature recognition prediction result at least includes the irregular body feature recognition result and the irregular body prediction result.

[0137] Optional, thermosphere state simulation submodule, specifically including:

[0138] a solar activity data unit, for which the third data includes at least solar activity parameters, geomagnetic disturbance parameters, and time parameters;

[0139] The model prediction unit is used to obtain the thermospheric dynamics simulation model based on TIEGCM, input solar activity parameters, geomagnetic disturbance parameters and time parameters into the thermospheric dynamics simulation model, and predict the thermospheric electric field and potential through the thermospheric dynamics simulation model;

[0140] The prediction result acquisition unit is used to obtain the thermosphere state prediction result from the thermosphere dynamics simulation model.

[0141] Optional, middle atmosphere modeling submodule, specifically including:

[0142] The middle and upper atmospheric parameter data unit is used for the fourth data including at least solar activity parameters, geomagnetic activity index, occultation atmospheric temperature and occultation atmospheric density;

[0143] The model prediction unit is used to obtain the trained Light-GBM mid- and high-level atmospheric density and temperature model, and input solar activity parameters, geomagnetic activity index, occultation atmospheric temperature, and occultation atmospheric density into the mid- and high-level atmospheric density and temperature model to predict the near-space environment through the mid- and high-level atmospheric density and temperature model.

[0144] The prediction result acquisition unit is used to obtain the atmospheric density and temperature prediction results from the middle and upper atmospheric density and temperature model.

[0145] Optional visualization and presentation module, including:

[0146] The data partition processing submodule is used to partition and organize the ionospheric storm prediction results, irregular body feature prediction results, thermosphere state prediction results, and atmospheric density and temperature prediction results according to spatial regions and time scales to obtain the organized prediction layer data;

[0147] The layer mapping submodule is used to map the predicted layer data to the preset product layers, which include the ionosphere layer, irregular body disturbance layer, thermosphere layer and middle and upper atmosphere layer;

[0148] The dynamic view synthesis submodule performs layer synthesis processing on the product layers based on the preset layer display rules to generate a visual mid- and high-level atmospheric forecast product view.

[0149] The specific definition of the mid- and upper-level atmospheric prediction system can be found in the definition of the mid- and upper-level atmospheric prediction method described above and will not be repeated here. Each module in the above-described mid- and upper-level atmospheric prediction system can be implemented in whole or in part through software, hardware, or a combination thereof. Each of the above-described modules can be embedded in or independent of a processor in a computer device in hardware form, or can be stored in a memory in a computer device in software form, so that the processor can call and execute the corresponding operations of each of the above modules.

[0150] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 9 As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data information such as multi-source space environment observation data, ionospheric storm prediction models, irregular body disturbance identification models, thermosphere dynamics simulation models and middle atmosphere density and temperature models. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a method for predicting the middle and upper atmosphere is implemented.

[0151] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the following steps are performed:

[0152] Obtain observation data from multiple sources of spatial environments and call the corresponding prediction model through the preset model scheduling strategy;

[0153] Based on the category of the prediction model, the corresponding target input data is matched in the observation data, and the target input data is input into the prediction model to obtain the prediction result;

[0154] Based on the prediction results, a visual mid- and upper-level atmosphere prediction product view is constructed, and the visual mid- and upper-level atmosphere prediction product view is presented through the display interface of the mid- and upper-level atmosphere prediction platform.

[0155] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0156] Obtain observation data from multiple sources of spatial environments and call the corresponding prediction model through the preset model scheduling strategy;

[0157] Based on the category of the prediction model, the corresponding target input data is matched in the observation data, and the target input data is input into the prediction model to obtain the prediction result;

[0158] Based on the prediction results, a visual mid- and upper-level atmosphere prediction product view is constructed, and the visual mid- and upper-level atmosphere prediction product view is presented through the display interface of the mid- and upper-level atmosphere prediction platform.

[0159] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the above-described method embodiments. Any reference to memory, storage, database, or other media used in the various embodiments provided herein may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM).

[0160] Those skilled in the art will clearly understand that for the sake of convenience and brevity in description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above.

[0161] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. A method for predicting the middle and upper atmosphere, characterized in that: Applied to a mid- and high-level atmosphere prediction platform, the mid- and high-level atmosphere prediction method includes: Obtain observation data from multiple sources of spatial environments and call the corresponding prediction model through the preset model scheduling strategy; Based on the category of the prediction model, matching corresponding target input data in the observation data, inputting the target input data into the prediction model to obtain a prediction result; Based on the prediction results, a visual mid- and high-level atmosphere prediction product view is constructed, and the visual mid- and high-level atmosphere prediction product view is presented through a display interface of the mid- and high-level atmosphere prediction platform.

2. The method for predicting the middle and upper atmosphere according to claim 1, wherein: The method of matching corresponding target input data from the observed data based on the category of the prediction model, inputting the target input data into the prediction model, and obtaining a prediction result specifically includes: When the category of the prediction model is an ionospheric storm prediction model, the corresponding target input data matched is the first data, and the obtained prediction result is the ionospheric storm prediction result; When the category of the prediction model is an irregular body feature recognition prediction model, the corresponding target input data is matched as the second data, and the obtained prediction result is an irregular body feature recognition prediction result; When the category of the prediction model is a thermospheric dynamics simulation model, the corresponding target input data matched is the third data, and the obtained prediction result is a thermospheric state prediction result; When the category of the prediction model is a mid-to-high-level atmospheric density and temperature model, the corresponding target input data matched is the fourth data, and the obtained prediction result is an atmospheric density and temperature prediction result.

3. The method for predicting the middle and upper atmosphere according to claim 2, characterized in that: When the category of the prediction model is an ionospheric storm prediction model, the corresponding target input data is matched as the first data, and the obtained prediction result is the ionospheric storm prediction result, which specifically includes: The first data at least includes ionospheric total electron content data, AE index and F10.7 solar activity index; The ionospheric total electron content data, the AE index, and the F10.7 solar activity index are input into the trained LSTM-CNN hybrid ionospheric storm prediction model to perform time series prediction through the ionospheric storm prediction model; The ionospheric storm prediction result is obtained from the ionospheric storm prediction model.

4. The method for predicting the middle and upper atmosphere according to claim 2, wherein: When the category of the prediction model is an irregular body feature recognition prediction model, the corresponding target input data is matched as the second data, and the obtained prediction result is an irregular body feature recognition prediction result, which specifically includes: The second data includes at least GNSS observation data, optical observation data, ionospheric detection parameters and radar detection data; Based on the pre-acquired edge extraction algorithm, phase coherence analysis algorithm, and machine learning algorithm, the irregular body feature recognition and prediction model is constructed, and the GNSS observation data, the optical observation data, the ionospheric sounding parameters, and the radar detection data are input into the irregular body feature recognition and prediction model to perform irregular body recognition and short-, medium-, and long-term forecasts through the irregular body feature recognition and prediction model; The irregular body feature recognition and prediction result is obtained from the irregular body feature recognition and prediction model, and the irregular body feature recognition and prediction result at least includes an irregular body feature recognition result and an irregular body prediction result.

5. The method for predicting the middle and upper atmosphere according to claim 2, wherein: When the category of the prediction model is a thermospheric dynamics simulation model, the corresponding target input data is matched as the third data, and the obtained prediction result is a thermospheric state prediction result, which specifically includes: The third data at least includes solar activity parameters, geomagnetic disturbance parameters and time parameters; Obtaining a thermospheric dynamics simulation model based on TIEGCM, inputting the solar activity parameter, the geomagnetic disturbance parameter, and the time parameter into the thermospheric dynamics simulation model, so as to predict the thermospheric electric field and electric potential through the thermospheric dynamics simulation model; The thermospheric state prediction result is obtained from the thermospheric dynamics simulation model.

6. The method for predicting the middle and upper atmosphere according to claim 2, characterized in that: When the category of the prediction model is a mid-to-high-level atmospheric density and temperature model, the corresponding target input data is matched as the fourth data, and the obtained prediction result is an atmospheric density and temperature prediction result, specifically including: The fourth data at least includes solar activity parameters, geomagnetic activity index, occultation atmosphere temperature and occultation atmosphere density; Obtaining the trained Light-GBM mid- and high-level atmospheric density and temperature model, and inputting the solar activity parameters, geomagnetic activity index, occultation atmospheric temperature, and occultation atmospheric density into the mid- and high-level atmospheric density and temperature model, so as to predict the near-space environment using the mid- and high-level atmospheric density and temperature model; The atmospheric density and temperature prediction results are obtained from the mid- and high-level atmospheric density and temperature model.

7. The method for predicting the middle and upper atmosphere according to claim 2, characterized in that: The step of constructing a visual mid- and upper-atmosphere forecast product view based on the forecast results specifically includes: Organizing the ionospheric storm prediction results, the irregular body feature prediction results, the thermosphere state prediction results, and the atmospheric density and temperature prediction results according to spatial regions and time scales to obtain organized prediction layer data; Mapping the predicted layer data to preset product layers, wherein the product layers include an ionosphere layer, an irregular body disturbance layer, a thermosphere layer, and a mid- and high-level atmosphere layer; Based on preset layer display rules, the product layer is subjected to layer synthesis processing to generate the visualized mid- and upper-level atmosphere forecast product view.

8. A mid- and upper-level atmospheric prediction system, characterized in that: The mid- and upper-level atmosphere prediction system comprises: The data acquisition management module is used to obtain observation data from multiple sources of spatial environments and call the corresponding prediction model through the preset model scheduling strategy; A prediction result generation module is used to match corresponding target input data in the observation data based on the category of the prediction model, input the target input data into the prediction model, and obtain a prediction result; The visualization and presentation module is used to construct a visualized mid- and high-level atmosphere prediction product view based on the prediction results, and present the visualized mid- and high-level atmosphere prediction product view through the display interface of the mid- and high-level atmosphere prediction platform.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method for predicting the middle and upper atmosphere according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for predicting the middle and upper atmosphere according to any one of claims 1 to 7 are implemented.

Images