Low-altitude meteorological multi-source data processing method, equipment and medium

Abstract

The invention provides a low-altitude meteorological multi-source data processing method and device and a medium, and belongs to the technical field of low-altitude meteorology. Meteorological data are collected in real time based on a ground station, a satellite, a radar and an unmanned aerial vehicle, accessed through multiple protocols and sorted; a standardized data set is generated through cleaning and normalization, and space coordinates are stored in a classified mode and associated; three-dimensional grid forecasting, severe convection early warning and disaster recognition products are generated through distributed calculation; the visualization engine dynamically displays a chart map and supports dragging interaction configuration; triggering an early warning rule to be issued to the aircraft through multiple channels; targeted early warning release is based on rule triggering and space matching, and the pertinence and the delivery rate of early warning are improved in combination with a multi-channel release and retry mechanism. The early warning effect is quantified through feedback data and live comparison, the acquisition strategy, the cleaning rule and the early warning threshold are continuously optimized, the system performance is dynamically improved, and the flight risk caused by meteorological factors is effectively reduced.

Description

Technical Field

[0001] This invention belongs to the field of low-altitude meteorological technology, specifically relating to a method, device, and medium for processing low-altitude meteorological multi-source data. Background Technology

[0002] With the deepening of low-altitude airspace management reform and the rapid development of the low-altitude economy, low-altitude flight activities such as unmanned aerial vehicle (UAV) logistics, urban air traffic (UAM), eVTOL (electric vertical take-off and landing aircraft), emergency rescue, and agricultural plant protection are becoming increasingly frequent, creating an urgent need for refined, timely, and reliable meteorological services.

[0003] Because low-altitude meteorological data involves a large volume of information, including observations, image files, and configuration information, packaged storage can easily lead to slow data read / write operations, time-consuming data retrieval, and an inability to quickly query meteorological data for a specific airspace. Related numerical weather prediction models target the mid-to-high altitudes, exhibiting low accuracy in forecasting small-scale weather phenomena such as low-altitude turbulence and localized severe convection. They struggle to identify complex features like severe convection, and their low-altitude meteorological outputs are mostly limited to temperature, wind speed, etc., which cannot meet the needs of low-altitude flight.

[0004] Existing displays are mostly fixed tables or static maps, which are insufficient to intuitively present the spatiotemporal changes of meteorological elements; moreover, they lack user-customizable configuration functions, failing to adapt to the viewing needs of different scenarios, resulting in low data utilization. Current warnings are mostly broadcast across the entire region, leading to interference from irrelevant warnings and affecting their timeliness. Summary of the Invention

[0005] This invention provides a method for processing low-altitude meteorological multi-source data. The method enables multi-source data acquisition, forecast generation, visualization, and precise targeted early warning, thereby improving the accuracy and reliability of low-altitude meteorological services.

[0006] The methods include: S101: Collects multi-source meteorological data, including real-time data from ground observation stations, low-altitude observation networks, satellites, radar, and UAV sensors; S102: Clean and convert the format of the collected multi-source meteorological data, remove redundant information and normalize it to generate a standardized dataset; S103: Store the standardized dataset in the database, classify and store it, and establish spatial coordinate system association; S104: Based on standardized stored data, low-altitude meteorological forecast data is generated through a distributed computing framework, including three-dimensional grid forecasts, severe convective weather warnings, and disaster identification results; S105: Displays low-air meteorological data from weather forecasts in real time in the form of charts and maps through a visualization canvas engine, and supports drag-and-drop interactive configuration of component layout and data binding; S106: Trigger warning rules based on real-time meteorological data and forecast low-altitude meteorological data, generate targeted warning information and release it to the aircraft through multiple channels; S107: Collect early warning response data and user feedback, combine real-time data to retrospectively analyze the accuracy of early warnings, and generate an evaluation report; S108: Optimize data collection strategies and early warning thresholds based on the evaluation report.

[0007] It should be further explained that S101 specifically includes the following steps: Connect to the data source server via FTP protocol at preset time intervals to download data files in the specified format; Establish a secure transfer connection using an SFTP client and complete the data file transfer using encryption algorithms; Configure an HTTP client to parse webpage content, extract meteorological information fields, and aggregate data; The collected data is sorted according to data category and feature parameters, a classification index is established, and the data is assigned to the processing queue. Configure data acquisition environment parameters, including transmission method selection and permission allocation settings.

[0008] It should be further explained that S103 specifically includes the following steps: Design a standardized table structure based on a relational database, define field types, lengths, and indexes, and create database tables; The data is divided into fixed-size blocks, each block is assigned a unique identifier and stored on distributed nodes. The mapping relationship between data blocks and files is recorded through a metadata server. To establish a spatial coordinate system association for low-altitude meteorological data, Geographic Information System (GIS) technology is used to bind the geographic coordinates of each data point with meteorological observation values ​​and store them as spatial data types. Configure a scheduled backup strategy to back up data to an off-site data center.

[0009] It should be further explained that S104 specifically includes the following steps: Extract standardized data and perform preprocessing before computation to generate the input dataset; The computing tasks are split based on a distributed computing framework, and subtasks are assigned to computing nodes. Run multi-source data fusion, assimilation, and forecasting algorithms on computing nodes to generate weather forecast results; Aggregate distributed computing results to generate standardized forecast low-altitude meteorological data; Verify the accuracy of forecast data and store forecast low-altitude meteorological data after adjusting for historical errors.

[0010] It should be further explained that S105 specifically includes the following steps: Initialize the visualization canvas engine, create a canvas container instance, configure the canvas size, coordinate system and rendering mode, and establish a binding connection with the browser rendering engine; Load the visualization component template from the component repository, instantiate the chart component and configure the basic properties of the component, including size, position and hierarchy, to complete the pre-layout of the component on the canvas; Configure data mapping rules, establish a connection channel between the component data binding interface and the backend data service, and set data refresh strategies and data format conversion rules; Perform canvas layout calculations, apply adaptive layout algorithms based on component type and quantity, adjust component position and size, and set canvas scaling ratio and viewport control parameters; Start the real-time rendering engine, establish a WebSocket long connection to receive real-time data, trigger the component redraw process, and execute dynamic updates and visual feedback for graphical elements.

[0011] It should be further explained that S106 specifically includes the following steps: Configure early warning trigger conditions, including element thresholds, early warning level, applicable airspace range, applicable aircraft type, and associated response measures; Real-time monitoring of meteorological data and matching it with early warning triggering conditions; recording the rule identifiers and element values ​​of successful matches. Based on the rules of successful matching, targeted early warning objects are filtered and a list of objects containing aircraft identification and location information is generated. The system calls the early warning information template to generate standardized early warning messages, fills in real-time data by variable substitution, and adds metadata. Select a release channel based on the target object list to send early warning information, monitor the reception status, and generate a release status report.

[0012] It should be further explained that S107 specifically includes the following steps: Configure multi-source data acquisition interfaces, establish connection channels with data systems of the regulatory end, the operation end, and the public end, and define data extraction rules and transmission protocols; Perform feedback data extraction operations to obtain early warning response records, user feedback data and real-time observation data in batches from the connection channel, and perform time alignment and format standardization. According to the scheduling plan, data retrieval or data push is performed, and the raw data is stored in the temporary storage area; The early warning data is spatiotemporally matched with the real-time data to calculate evaluation indicators such as hit rate, false alarm rate, and false alarm rate. Generate an evaluation and analysis report, integrate evaluation indicator results and feedback data, and output the report document.

[0013] It should be further explained that S108 specifically includes the following steps: Define the evaluation indicator system and calculate the indicator values, and generate an indicator summary table; An evaluation report is generated based on the indicator summary table, including problem identification and root cause analysis; Adjust the data acquisition strategy and data preprocessing process based on the root causes of the problems identified in the assessment report; Update model parameters and early warning thresholds based on the adjusted data acquisition strategy; Verify the effectiveness of the optimized system, and solidify the strategy or re-evaluate based on the verification results.

[0014] According to another embodiment of this application, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the low-altitude meteorological multi-source data processing method.

[0015] According to another embodiment of this application, a storage medium is also provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the low-altitude meteorological multi-source data processing method.

[0016] As can be seen from the above technical solutions, the present invention has the following advantages: This invention provides a low-altitude meteorological multi-source data processing method that collects and processes multi-dimensional meteorological data from the ground, low altitude, satellite, radar, and UAVs, eliminating redundant and abnormal data and unifying formats, units, and coordinates. It selects appropriate storage methods based on data characteristics and achieves precise binding of data to geographical locations through spatial indexing, improving data retrieval and spatial analysis efficiency. The fusion of multi-source data assimilation and numerical models enables accurate identification of complex weather conditions such as severe convection, achieving diverse outputs such as 3D grids, severe convection warnings, and disaster identification, meeting the needs of low-altitude flight. Visualization is supported through intuitive charts, maps, and interactive configurations, supporting multi-terminal adaptation and facilitating real-time viewing by ground command and aircraft. Targeted warning issuance is based on rule triggering and spatial matching, combined with multi-channel issuance and retry mechanisms to improve the targeting and delivery rate of warnings. By comparing feedback data with real-time conditions, the warning effect is quantified, and the collection strategy, cleaning rules, and warning thresholds are continuously optimized to achieve dynamic improvement in system performance, effectively reducing flight risks caused by meteorological factors. Attached Figure Description

[0017] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the description will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 Flowchart of low-altitude meteorological multi-source data processing method; Figure 2 This is a schematic diagram of an electronic device. Detailed Implementation

[0019] The low-altitude meteorological multi-source data processing method of this application will be described in detail below. Specific details such as particular system structures and technologies are presented for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application can also be implemented in other embodiments without these specific details.

[0020] The statements such as "one embodiment" or "some embodiments" described in this application mean that one or more embodiments of this application include the specific features, structures, or characteristics described in that embodiment. Therefore, the statements such as "in one embodiment," "in some embodiments," "in other embodiments," and "in still other embodiments" in this application do not necessarily refer to the same embodiment, but rather mean one or more, but not all, embodiments, unless otherwise specifically emphasized.

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Please see Figure 1 The diagram shows a flowchart of a low-altitude meteorological multi-source data processing method in a specific embodiment. The method includes: S101: Collects multi-source meteorological data, including real-time data from ground observation stations, low-altitude observation networks, satellites, radars, and UAV sensors, and accesses and sorts the data to preset categories via FTP / SFTP / HTTP protocols.

[0023] In some embodiments, real-time data of basic meteorological elements such as temperature, air pressure, humidity, wind speed and direction are obtained from ground observation stations; low-altitude observation networks can obtain meteorological information in low-altitude areas, such as low-altitude wind fields and temperature stratification; satellites can provide large-scale meteorological images and monitoring data, such as cloud images and precipitation estimates; radar can monitor the location, intensity and movement of weather systems such as precipitation and storms in real time; and UAV sensors can flexibly collect meteorological data in specific areas, such as acquiring local temperature, humidity, and air quality data.

[0024] Optionally, S101 specifically includes the following steps: Step S1011: Establish an FTP connection according to a preset time interval, scan the data source server directory, and download data files in the specified format; users can adjust the interval parameters or trigger manual downloads through the interface to obtain missing data. This achieves schedulable data collection and ensures the continuity and real-time nature of data collection.

[0025] Step S1012: Configure transfer parameters using an SFTP client, verify user permissions using the SSH protocol, and execute file transfer; enable encryption algorithms during the transfer to ensure data security. This improves the security and reliability of data transmission.

[0026] Step S1013: Configure the HTTP client, parse the WebPage content and collect meteorological-related data according to the preset data configuration list, including dynamically extracting meteorological information fields from the webpage.

[0027] Step S1014: Sort the collected data according to data category, step size, frequency, size, and type, establish a data classification index, and allocate it to the corresponding queue.

[0028] This embodiment categorizes collected data according to preset data classification rules, such as meteorological observations and radar data, and processes the collected data accordingly. Logical judgments based on step size, frequency, file size, and type are used to assign data to the corresponding classification queue. This achieves orderly data organization and improves the efficiency of subsequent data processing.

[0029] Step S1015: Configure data acquisition environment parameters, including transmission method selection, transmission format definition, and transmission permission allocation, to ensure the security and compatibility of data transmission.

[0030] It should be noted that various meteorological observation devices use sensors to convert meteorological physical quantities into electrical signals, which are then digitized by a data acquisition unit and transmitted to a data processing center via network protocols. The data processing center classifies and stores the data according to preset rules. This protocol-based access and sorting makes data management more organized and facilitates processing.

[0031] S102: Clean and convert the format of the collected multi-source meteorological data, remove redundant information and normalize it to generate a standardized dataset.

[0032] In some embodiments, data cleaning involves checking for outliers, such as temperature and wind speed values ​​that significantly exceed normal ranges, and removing them by setting reasonable thresholds; handling missing values ​​by using interpolation or other methods to fill in missing data points; and removing duplicate data records. Format conversion involves converting different formats of data from different sources into a unified standard format defined by the system, such as converting a specific binary format output by certain devices into a common CSV or JSON format. Normalization involves scaling data to a specific range according to certain rules, such as scaling data to the [0,1] range, to make data of different dimensions comparable.

[0033] S103: Store the standardized dataset in the database, classify and store it, and establish spatial coordinate system association.

[0034] In some embodiments, S103 specifically includes the following steps: Step S1031: Design a standardized table structure based on a relational database, define field types, lengths and indexes, use UTF-8 encoding, and create a database table to store data.

[0035] Step S1032: Divide the data into fixed-size blocks, assign a unique identifier to each block, store them on distributed nodes, and record the mapping relationship between data blocks and files through a metadata server.

[0036] Step S1033: Establish a spatial coordinate system association for low-altitude meteorological data. Using Geographic Information System (GIS) technology, bind the geographic coordinates of each data point with meteorological observation values ​​and store them as spatial data types.

[0037] Step S1034: Set up a scheduled backup strategy to back up data to an off-site data center, using a combination of incremental backup and full backup to ensure data security.

[0038] Step S1035: Build a data query service based on SQL query language and GIS API.

[0039] It should be noted that step S1031 designs the table structure, defines field types and constraints based on the business characteristics and storage requirements of the data to ensure data integrity and consistency. Step S1032 divides large files into multiple data blocks, each stored on a different node, and records the correspondence between files and data blocks through a metadata server. When a file needs to be read, the client obtains the data block location information from the metadata server, then reads the data block from the corresponding node and restores it to the original file. This allows for efficient storage and management of data availability. Step S1033 utilizes GIS technology to bind geographic coordinates with meteorological observation values, storing them as a spatial data type. Spatial data can be quickly queried through indexes. Step S1034 periodically backs up the data to a designated location and restores the backup data to the database when needed. Step S1035 builds a data query service based on SQL query language and GIS API. The interface receives query requests, parses query conditions, generates corresponding SQL query statements or GIS query operations, executes the query, and returns the results to the client. This provides efficient data access services for applications and enables data sharing.

[0040] S104: Based on standardized stored data, low-altitude meteorological forecast data is generated through a distributed computing framework, including three-dimensional grid forecasts, severe convective weather warnings, and disaster identification results.

[0041] S104 of this embodiment specifically includes the following steps: S1041: Extract standardized data and perform preprocessing before calculation; filter standardized data from the database according to a preset time range; unify data from different collection frequencies to the same timestamp through time alignment; associate discrete station data with a preset grid, and supplement missing data within the grid through spatial interpolation; finally generate an input dataset containing time, longitude, latitude, and meteorological element values.

[0042] S1042: Based on a distributed computing framework, the computation task is split, and the input dataset is divided into spatial grid units and time segments to generate several sub-tasks; The cluster resource monitoring module collects the processing power and network bandwidth of each computing node, and allocates subtasks to nodes with lower loads; it also configures the dependent resources for each subtask and registers the task execution order in the task scheduler.

[0043] S1043: Runs a multi-source data fusion, assimilation, and forecasting algorithm on the computing node, calls the three-dimensional variational assimilation module to fuse the input data, and assimilates satellite, radar, and ground observation data into a unified analysis field; starts the numerical forecasting mode, solves the fluid dynamics equations based on the assimilated analysis field, and generates the spatial distribution of elements such as temperature, humidity, and wind speed for future time periods; identifies strong convective cloud clusters by extracting features such as radar echo intensity and vertical velocity, and combining them with a trained CNN model; runs the disaster identification module, and filters hazardous meteorological elements based on preset thresholds.

[0044] S1044: Aggregate distributed computing results and generate standardized products. After each computing node completes its subtask, it uploads the results to the master node. The master node stitches together the results of subtasks according to spatial grid and temporal order to form a regional time series dataset; the dataset is converted to a standard format using a format conversion tool; metadata is added to each low-altitude meteorological data to generate forecast low-altitude meteorological data.

[0045] S1045: Extract forecast data for a preset time period from the forecast low-altitude meteorological data and compare it with the actual data for the corresponding time period; calculate the error index; if the error exceeds the preset threshold, adjust the forecast low-altitude meteorological data based on the linear correction method of historical error statistics; after verification, classify and store the forecast low-altitude meteorological data.

[0046] In step S104 of this embodiment, preprocessing before calculation eliminates frequency differences, spatial dispersion, and format chaos in multi-source data through time alignment, spatial interpolation, and structured organization. Distributed task splitting combined with dynamic allocation of cluster resources can efficiently utilize the computing power of multiple nodes to process massive amounts of low-altitude meteorological data, meeting the timeliness requirements of low-altitude forecasts. Standardized format conversion and metadata addition during result aggregation ensure that the forecasted low-altitude meteorological data conforms to industry-standard specifications, facilitating direct access for visualization, early warning triggering, and other processes. Simultaneously, the metadata provides a basis for tracing the source of the forecasted low-altitude meteorological data, enhancing data credibility. Through comparison and verification of forecast data with actual data and linear correction, systematic errors can be corrected in a timely manner, further improving forecast accuracy and ensuring that the final output identification results accurately match the meteorological information related to low-altitude flight.

[0047] S1043 of this embodiment specifically includes the following steps: To address equipment errors in satellite, radar, and ground observation data, the deviation correction coefficients of each data source are statistically calculated based on historical matching data. The input satellite emissivity, radar reflectivity, and ground meteorological element values ​​are then corrected one by one to eliminate system errors across equipment.

[0048] For the multi-source data after bias correction, the spatial consistency of satellite cloud images, the rationality of vertical liquid water content in radar echoes, and the calibration frequency of ground station data, which reflect the reliability of the data, are calculated. The quality indicators are transformed into fusion weights, and candidate values ​​for the initial analysis field are generated through weighted fusion.

[0049] The error distribution of each data source is simulated using the Monte Carlo method, generating multiple sets of input samples with errors. These samples are then input into a three-dimensional variational assimilation module to obtain multiple analysis field results. The mean and variance of the analysis fields are statistically analyzed to quantify the prediction uncertainty of each grid point.

[0050] For thunderstorm cells and low-level wind shear zones in low-altitude meteorology, the background error covariance matrix of the three-dimensional variational assimilation is adjusted to increase the vertical background error variance of wind field and humidity in the 0-5km altitude layer, reduce the smoothing effect of large-scale background field on small-scale features, and preserve the vertical structural details of the convective system.

[0051] Based on the reasonable range of atmospheric stability, vertical wind shear, and temperature lapse rate of low-air meteorology, the assimilated analysis field is checked grid by grid: if the temperature lapse rate of a certain grid exceeds 10K / km, or the vertical wind shear exceeds 30m / s / s, the outlier value is replaced with the mean value of the neighboring grid or the model background field value to ensure that the analysis field conforms to the real atmospheric physical logic.

[0052] As can be seen, this embodiment defines specific indicators for each data source. For example, satellite data shows that higher cloud coverage indicates more reliable brightness temperature observations. Radar data shows that the variation of VIL with altitude conforms to the structure of convective clouds. Ground station data considers stations calibrated in the past 7 days. Quality scores are normalized and used as fusion weights. Based on the Monte Carlo method, the impact of data errors is simulated through random sampling. The mean of the analysis field represents the most likely true state, and the variance represents the degree of uncertainty of this state. Areas with large variance indicate large data errors, requiring caution in forecasting. Low-air meteorological hazards are mostly small- to medium-scale systems, and the large-scale background field smooths these small-scale features. By adjusting the B matrix, assimilation focuses more on small- to medium-scale changes, preserving the fine structure of convective systems. This embodiment considers the inherent laws of atmospheric physical quantities; for example, the temperature lapse rate of the stable atmosphere does not exceed 9.8 K / km. Assimilation or interpolation may lead to outliers, which can disrupt the model's physical processes. This ensures that the analysis field conforms to the true atmospheric physical logic, avoiding erroneous forecast results due to input outliers.

[0053] In some embodiments, the multi-source data fusion assimilation and prediction algorithm can employ the 3DVar variational assimilation model, which can be used to fuse satellite, radar, and ground observation data into a unified analysis field, and obtain the optimal analysis field by minimizing the differences between observations and models.

[0054] Alternatively, based on the Monte Carlo Data Assimilation Model (MCDA), multiple sets of error samples can be generated to simulate the impact of data errors and quantify the uncertainty of the field.

[0055] Step S105: Display the low-altitude meteorological data of the weather forecast in the form of charts and maps in real time through the visualization canvas engine, and support drag-and-drop interactive configuration of component layout and data binding.

[0056] In some embodiments, S105 specifically includes the following steps: Step S1051: Initialize the visualization canvas engine, create a canvas container instance, configure the canvas size, coordinate system and rendering mode, and establish a binding connection with the browser rendering engine.

[0057] In some embodiments, DOM element IDs and CSS style properties are set when the canvas container is created, and the WebGL or Canvas2D rendering context is initialized. The coordinate system is configured to use Cartesian or polar coordinates, and scaling and translation transformation matrices are set. In this embodiment, a rendering pipeline is established after the canvas engine is initialized, converting graphic elements into pixel data. Using layered rendering technology, the background, charts, and interactive elements are rendered to different layers respectively, and finally, the final image is composited.

[0058] Step S1052: Load the visualization component template from the component repository, instantiate the chart component and configure the basic properties of the component, including size, position and hierarchy, to complete the pre-layout of the component on the canvas.

[0059] In some embodiments, component template definition files are retrieved from the component repository, and component properties, including width, height, position coordinates, rotation angle, and opacity, are set during instantiation. A component tree data structure is established to manage parent-child and sibling relationships between components, improving component reusability and maintainability.

[0060] Step S1053: Configure data mapping rules, establish a connection channel between the component data binding interface and the backend data service, and set data refresh strategy and data format conversion rules.

[0061] In some embodiments, a data mapping rule table is configured to define the correspondence between component attributes and data fields. Data refresh strategies are set, including polling frequency, incremental update conditions, and full update trigger conditions. Data format conversion rules define the conversion logic from raw data to visualized data. Scalability is achieved by supporting multiple data adapters, ensuring compatibility with different data source formats.

[0062] Step S1054: Perform canvas layout calculation, apply adaptive layout algorithm according to component type and quantity, adjust component position and size, and set canvas scaling ratio and viewport control parameters.

[0063] In some embodiments, the layout engine traverses the component tree, calculating the desired position and size of each component. The layout algorithm iteratively optimizes until a convergence condition is met or the maximum number of iterations is reached. The final result is then passed to the rendering engine for drawing. This improves space utilization and visual effects; it also adapts to different screen sizes and display requirements.

[0064] Step S1055: Start the real-time rendering engine, establish a WebSocket long connection to receive real-time data, trigger the component redraw process, and perform dynamic updates and visual feedback for graphical elements.

[0065] This embodiment establishes a WebSocket long-lived connection to communicate with the data server. Upon arrival of real-time data, a data preprocessing process is triggered, including outlier filtering and data smoothing. The component redraw engine updates graphical attributes based on data changes and executes animation transition effects. The visual feedback system highlights changing data, providing visual cues. This embodiment provides a real-time data visualization experience; promptly reflects data changes and trends; and enhances immediate feedback for user interaction.

[0066] S106: Trigger warning rules based on real-time meteorological data and forecast low-altitude meteorological data, generate targeted warning information and release it to the aircraft through multiple channels.

[0067] S106 specifically includes the following steps: S1061: Configure early warning trigger conditions, classify and store rules according to meteorological element types, and each rule includes element threshold, early warning level, applicable airspace range, applicable aircraft type and associated response measures.

[0068] S1062: Real-time monitoring of data and rule matching, deployment of data listening process, extraction of short-term forecast values ​​of wind speed, visibility, precipitation intensity and forecast low-altitude meteorological data from real-time meteorological database; determination of whether trigger conditions are met by threshold comparison; for multi-element joint rules, such as joint prediction of wind speed and visibility, verification of whether all conditions are met simultaneously by logical AND operation, and recording the rule ID and corresponding element value of successful matching.

[0069] S1063: Filter target warning objects, connect to the low-altitude flight service platform database, obtain the real-time location, flight plan, registration model and communication method of currently active aircraft; based on the airspace range in the successfully matched rules, determine whether the aircraft is within the warning airspace through the spatial inclusion algorithm; combine the applicable aircraft types in the rules, filter out the aircraft that meet the conditions, and generate a list of target objects containing the aircraft ID, affiliated unit and current location.

[0070] S1064: Generate standardized early warning information, call the early warning information template library, and select the corresponding template according to the early warning level and feature type of the matching rules; the template contains the early warning type, and the real-time feature value, early warning airspace, and suggested measures are filled into the placeholders through the variable substitution algorithm; after the information is generated, the rule ID, release timestamp and digital signature are added to form the final early warning message.

[0071] S1065: Select the corresponding publishing channel according to the communication method of the aircraft in the target object list; after publishing, listen to the acknowledgment signal of each channel, record the receiving status of each aircraft, retry the sending of failed sending tasks, and finally generate a publishing status report and store it in the log database.

[0072] It should be noted that the electronic device involved in this embodiment has a quick query port for historical service products on its display interface, enabling customized queries of processed and published low-air meteorological service products based on combinations of multiple elements such as region, time, product category, industry, weather process, and service topic. The electronic device can monitor meteorological service products in real time, achieving real-time monitoring of the release status of low-air meteorological service products and enabling front-end categorized display of published products.

[0073] Electronic equipment monitors the basic data, monitoring, forecasting, and early warning data, interfaces, and services used by low-altitude meteorological service products, displays them in a list, determines whether they are acquired on time, and issues alarms for abnormal situations.

[0074] The electronic equipment can provide accurate and real-time low-altitude meteorological information to ensure low-altitude flight safety. By monitoring key meteorological elements such as wind direction, wind speed, temperature, humidity, and visibility in real time, and combining historical meteorological data with current environmental trends, the system can intelligently predict potential meteorological risks, providing scientific decision-making support for low-altitude management departments. The system has an efficient targeted early warning information dissemination and emergency response mechanism. Once adverse weather conditions occur, it can quickly activate the early warning process to ensure the safe operation of aircraft.

[0075] The electronic equipment is equipped with a digital intelligent low-altitude meteorological safety assurance platform, which consists of four parts: a low-altitude meteorological forecasting subsystem, a low-altitude meteorological monitoring subsystem, a low-altitude meteorological operation subsystem, and a low-altitude meteorological public service subsystem.

[0076] The low-altitude meteorological forecasting subsystem utilizes low-altitude meteorological data from the low-altitude meteorological big data system, applying numerical models, fluid dynamics, and AI large-scale models to generate three-dimensional low-altitude meteorological forecast products that meet low-altitude requirements. Through analysis of different airspaces and different types of eVTOL and UAVs, it provides refined and personalized early warning services. Employing technologies such as message flow, natural language processing models, and BeiDou communication, it develops core functions such as intelligent matching, automatic generation, and intelligent quality control of early warning information. Combined with methods such as low-altitude fencing and electronic fencing, it achieves precise early warning information dissemination for different airspaces and different aircraft. The system supports the collection of feedback data from regulatory, operational, and public ends, as well as the integration of real-time data, to assess the accuracy of low-altitude forecasts and early warnings, continuously improving the accuracy and timeliness of low-altitude meteorological forecast services.

[0077] The low-altitude meteorological supervision subsystem monitors and analyzes comprehensive weather conditions across the entire area, and combines this with precise meteorological early warning information to promptly send detailed and clear safety and emergency requirements to drone take-off and landing sites, commercial low-altitude flight destinations, and operational teams carrying out flight missions within the supervised area. This enables the classification, tiered, and targeted dissemination of meteorological early warning information, ensuring the timely detection and handling of meteorological risks during flight, and providing targeted safety guidance and emergency response measures, thereby effectively reducing the incidence of low-altitude flight accidents.

[0078] The low-altitude meteorological operations subsystem enhances the meteorological disaster response capabilities of low-altitude air traffic operators by aggregating data from self-built observation equipment and synchronizing observation data and forecast products from meteorological bureaus. It enables real-time monitoring and early warning of meteorological disasters at takeoff and landing sites, along flight routes, and during flight, ensuring accurate and timely meteorological information for low-altitude flights. The system also possesses the capability for joint prevention and control with regulatory authorities and meteorological bureaus, as well as the ability to synchronize disaster early warning information, thereby providing comprehensive meteorological support services for low-altitude flights.

[0079] The Low-Altitude Meteorological Public Service Subsystem primarily targets small businesses, teams, and individuals in the low-altitude airspace. It provides low-altitude meteorological forecast queries and general service product inquiries, supporting queries for real-time and forecast data on locations and flight paths of interest via web pages, APIs, and mini-programs. It also provides access to all warning information issued by the meteorological bureau and regulatory platforms, ensuring users are promptly informed of meteorological risks that may affect low-altitude flight safety. Through AI-powered question-and-answer services, the system offers personalized meteorological information and suggestions to meet the needs of diverse users, addressing the inconvenience of accessing meteorological information and insufficient risk awareness that may arise for the public during low-altitude flights, thus providing strong public support for low-altitude flight safety.

[0080] S107: Collect early warning response data and user feedback, combine real-time data to retrospectively analyze the accuracy of early warnings, and generate an assessment report.

[0081] In some embodiments, S107 specifically includes the following steps: Step S1071: Configure the multi-source data acquisition interface, establish connection channels with the data systems of the regulatory end, the operation end, and the public end, and define data extraction rules and transmission protocols.

[0082] Step S1072: Perform feedback data extraction operation, obtain early warning response records, user feedback data and real-time observation data in batches from the connection channel, and perform time alignment and format standardization.

[0083] This embodiment executes data retrieval or data push according to the scheduling plan, storing the raw data in a temporary storage area. The data processor applies cleaning rules and quality checks to mark problematic data. The time alignment module uses interpolation algorithms to fill time gaps, ensuring data temporal consistency. The format converter maps the data to a standard structure for subsequent analysis.

[0084] Step S1073: Perform spatiotemporal matching between the early warning data and the real-time data, and calculate evaluation indicators such as hit rate, missed rate, and false alarm rate.

[0085] In some embodiments, spatiotemporal matching of data is performed to associate each early warning event with real-time observations within the corresponding spatiotemporal range. Then, a confusion matrix is ​​constructed based on the matching results, and various evaluation indicators are calculated. A statistical testing module analyzes the significance of the indicator values ​​and identifies systematic biases. The results analyzer generates in-depth analysis conclusions, identifying early warning performance patterns and areas for improvement.

[0086] Step S1074: Generate an evaluation and analysis report, integrate the evaluation indicator results and feedback data, and output the report document.

[0087] Step S1075: Establish an evaluation result feedback link, distribute the evaluation report to relevant system modules, and update the early warning rule parameters and threshold settings.

[0088] In some embodiments, evaluation report distribution rules are configured, defining the receiving system, distribution timing, and transmission method. Early warning thresholds and rule parameters are automatically adjusted based on the evaluation results. System configuration version control is implemented, recording parameter change history and effect tracking. Feedback monitoring is established to track the actual effects of optimization measures.

[0089] S108: Optimize data collection strategies and early warning thresholds based on the evaluation report.

[0090] In some embodiments, S108 specifically includes the following steps: S1081: Define the evaluation index system and calculate the index values, and generate an index summary table.

[0091] Specifically, the evaluation index system includes: early warning accuracy rate (the number of correct early warnings divided by the total number of early warnings); hit rate (the number of correctly warned and actually occurring disasters divided by the actual number of disasters); F1 score (calculated as 2 multiplied by the early warning accuracy rate multiplied by the hit rate, then divided by the sum of the early warning accuracy rate and the hit rate; and user satisfaction score (based on a questionnaire survey, ranging from 1 to 5 points, with the average score taken), covering aircraft operators and low-altitude airspace management departments. Based on three months of historical early warning records, including early warning time, type, and scope, as well as actual disaster occurrence logs containing disaster time, type, and affected area, the Python pandas library was used to calculate the values ​​of each index, and a summary table of the indices was output.

[0092] S1082: Generate an evaluation report based on the indicator summary table, including problem identification and root cause analysis.

[0093] Specifically, the report includes an overview of indicators, i.e., a table of indicators for each disaster type; problem identification, highlighting disaster types with poor early warning effects, such as a 65% hit rate for thunderstorms and strong winds; and root cause analysis, linking data collection, model prediction, and threshold setting processes, for example, the low hit rate for thunderstorms and strong winds was due to a 15-minute delay in satellite CTH data, causing the model's initial field to fail to capture early convective characteristics. The report is output in PDF or Excel format and includes trend graphs of key indicators, such as the hit rate change curve over the past six months.

[0094] S1083: Adjust the data acquisition strategy and data preprocessing process based on the root causes of the problems in the assessment report.

[0095] Specifically, data acquisition will be adjusted based on the root cause of the problem in S1082. If the delay is due to satellite data, the frequency of CTH data acquisition from the Fengyun-4 satellite will be increased from once every 10 minutes to once every 5 minutes. At the same time, a caching mechanism will be added to the local receiving server to reduce transmission time. If the delay is due to missing ground station data, 10 additional automatic low-altitude stations in key areas will be added to collect temperature, humidity, and wind field data and connect them to the existing data pipeline. The data preprocessing process will be adjusted, and satellite data timestamp calibration will be added to eliminate delays caused by time zone deviations.

[0096] S1084: Update model parameters and early warning thresholds based on the adjusted data acquisition strategy.

[0097] Specifically, using disaster cases from the past 6 months, including 100 thunderstorm and gale cases and 50 low-level wind shear cases, the LightGBM disaster potential prediction model was retrained. The hyperparameters were adjusted, increasing the tree depth from 5 to 8 and decreasing the learning rate from 0.1 to 0.05 to capture more complex small- and medium-scale features. In combination with business needs, the warning threshold for thunderstorm and gale was lowered from 70% to 65%. At the same time, the false alarm rate was monitored using data from the past month. If the false alarm rate exceeded 15%, it was adjusted back to 68% to balance the hit rate and the false alarm rate.

[0098] S1085: Verify the effectiveness of the optimized system, and solidify the strategy or re-evaluate based on the verification results.

[0099] Specifically, the new strategy will be implemented for one month, collecting early warning records and actual disaster logs. The indicators in S1081 will be recalculated, such as increasing the hit rate of thunderstorms and strong winds from 65% to 75% and reducing the false alarm rate from 35% to 25%. A new evaluation report will be generated to compare the changes in indicators before and after optimization. If the effect is satisfactory, i.e., the hit rate is greater than or equal to 70% and the false alarm rate is less than or equal to 30%, the optimization strategy will be solidified into the system, such as updating the data collection frequency configuration and model version. If the effect is not satisfactory, the process will return to S1082 to re-identify the problem.

[0100] It should be understood that the sequence number of each step in the above embodiments does not imply 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 the present invention.

[0101] like Figure 2 As shown, this application also provides an electronic device, including a display module 103, a memory 102, a processor 101, a communication module 104, and a computer program stored in the memory and executable on the processor 101. When the processor 101 executes the program, it implements the steps of a low-altitude meteorological multi-source data processing method.

[0102] In embodiments of the present invention, electronic devices include, but are not limited to, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic devices may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the embodiments described and / or claimed herein.

[0103] In this embodiment, processor 101 may be implemented using at least one of an application-specific integrated circuit, a programmable logic device, a field-programmable gate array, a processor, a controller, a microcontroller, a microprocessor, or an electronic unit designed to perform the functions described herein. In some cases, such an implementation may be implemented within a controller. For software implementation, implementations such as processes or functions may be implemented with separate software modules that allow the performance of at least one function or operation. Software code may be implemented by a software application (or program) written in any suitable programming language, and the software code may be stored in memory and executed by the controller.

[0104] The display module 103 is used to display information input by the user or information provided to the user. The display module 103 may include a display panel, which may be configured in the form of a liquid crystal display, an organic light-emitting diode, or the like.

[0105] The memory 102 can be used to store software programs and various data. The memory 102 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0106] The communication module 104 transmits radio signals to and / or receives radio signals from at least one of a base station, an external terminal, and a server. Such radio signals may include voice call signals, video call signals, or various types of data sent and / or received according to text and / or multimedia messages.

[0107] The present invention also provides a storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the low-altitude meteorological multi-source data processing method.

[0108] The storage medium may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example,, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of readable storage media include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0109] The storage medium stores a program product capable of implementing the methods described above in this specification. In some possible implementations, various aspects of this disclosure can also be implemented as a program product comprising program code that, when run on a terminal device, causes the terminal device to perform the steps described in the exemplary methods section of this specification according to various exemplary embodiments of this disclosure.

[0110] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for processing low-altitude meteorological multi-source data, characterized in that, The methods include: S101: Collects multi-source meteorological data, including real-time data from ground observation stations, low-altitude observation networks, satellites, radar, and UAV sensors; S102: Clean and convert the format of the collected multi-source meteorological data, remove redundant information and normalize it to generate a standardized dataset; S103: Store the standardized dataset in the database, classify and store it, and establish spatial coordinate system association; S104: Based on standardized stored data, low-altitude meteorological forecast data is generated through a distributed computing framework, including three-dimensional grid forecasts, severe convective weather warnings, and disaster identification results; S105: Displays low-air meteorological data from weather forecasts in real time in the form of charts and maps through a visualization canvas engine, and supports drag-and-drop interactive configuration of component layout and data binding; S106: Trigger warning rules based on real-time meteorological data and forecast low-altitude meteorological data, generate targeted warning information and release it to the aircraft through multiple channels; S107: Collect early warning response data and user feedback, combine real-time data to retrospectively analyze the accuracy of early warnings, and generate an evaluation report; S108: Optimize data collection strategies and early warning thresholds based on the evaluation report.

2. The low-altitude meteorological multi-source data processing method according to claim 1, characterized in that, S101 specifically includes the following steps: Connect to the data source server via FTP protocol at preset time intervals to download data files in the specified format; Establish a secure transfer connection using an SFTP client and complete the data file transfer using encryption algorithms; Configure an HTTP client to parse webpage content, extract meteorological information fields, and aggregate data; The collected data is sorted according to data category and characteristic parameters, a classification index is established, and the data is assigned to the processing queue.

3. The low-altitude meteorological multi-source data processing method according to claim 1, characterized in that, S103 specifically includes the following steps: Design a standardized table structure based on a relational database, define field types, lengths, and indexes, and create database tables; The data is divided into fixed-size blocks, each block is assigned a unique identifier and stored on distributed nodes. The mapping relationship between data blocks and files is recorded through a metadata server. To establish a spatial coordinate system association for low-altitude meteorological data, Geographic Information System (GIS) technology is used to bind the geographic coordinates of each data point with meteorological observation values ​​and store them as spatial data types. Configure a scheduled backup strategy to back up data to an off-site data center.

4. The low-altitude meteorological multi-source data processing method according to claim 1, characterized in that, S104 specifically includes the following steps: S1041: Extract standardized data and perform preprocessing before computation to generate a structured input dataset; S1042: Based on a distributed computing framework, computing tasks are split and subtasks are allocated to computing nodes; S1043: Run multi-source data fusion, assimilation, and forecasting algorithms on computing nodes to generate weather forecast results; S1044: Aggregate distributed computing results to generate standardized forecast products; S1045: Verify the accuracy of forecast data and store forecast low-altitude meteorological data after adjusting for historical errors.

5. The low-altitude meteorological multi-source data processing method according to claim 1, characterized in that, S105 specifically includes the following steps: Initialize the visualization canvas engine, create a canvas container instance, configure the canvas size, coordinate system and rendering mode, and establish a binding connection with the browser rendering engine; Load the visualization component template from the component repository, instantiate the chart component and configure the basic properties of the component, including size, position and hierarchy, to complete the pre-layout of the component on the canvas; Configure data mapping rules, establish a connection channel between the component data binding interface and the backend data service, and set data refresh strategies and data format conversion rules; Perform canvas layout calculations, apply adaptive layout algorithms based on component type and quantity, adjust component position and size, and set canvas scaling ratio and viewport control parameters; Start the real-time rendering engine, establish a WebSocket long connection to receive real-time data, trigger the component redraw process, and execute dynamic updates and visual feedback for graphical elements.

6. The low-altitude meteorological multi-source data processing method according to claim 1, characterized in that, S106 specifically includes the following steps: Configure early warning trigger conditions, including element thresholds, early warning level, applicable airspace range, applicable aircraft type, and associated response measures; Real-time monitoring of meteorological data and matching it with early warning triggering conditions; recording the rule identifiers and element values ​​of successful matches. Based on the rules of successful matching, targeted early warning objects are filtered and a list of objects containing aircraft identification and location information is generated. The system calls the early warning information template to generate standardized early warning messages, fills in real-time data by variable substitution, and adds metadata. Select a release channel based on the target object list to send early warning information, monitor the reception status, and generate a release status report.

7. The low-altitude meteorological multi-source data processing method according to claim 1, characterized in that, S107 specifically includes the following steps: Configure multi-source data acquisition interfaces, establish connection channels with data systems of the regulatory end, the operation end, and the public end, and define data extraction rules and transmission protocols; Perform feedback data extraction operations to obtain early warning response records, user feedback data and real-time observation data in batches from the connection channel, and perform time alignment and format standardization. According to the scheduling plan, data retrieval or data push is performed, and the raw data is stored in the temporary storage area; The early warning data is spatiotemporally matched with the real-time data to calculate evaluation indicators such as hit rate, false alarm rate, and false alarm rate. Generate an evaluation and analysis report, integrate evaluation indicator results and feedback data, and output the report document.

8. The low-altitude meteorological multi-source data processing method according to claim 1, characterized in that, S1043 specifically includes the following steps: To address equipment errors in satellite, radar, and ground observation data, the deviation correction coefficients of each data source are statistically calculated based on historical matching data. The input satellite emissivity, radar reflectivity, and ground meteorological element values ​​are then corrected one by one to eliminate systematic errors across equipment. For the multi-source data after bias correction, the spatial consistency of satellite cloud images, the rationality of vertical liquid water content in radar echoes, and the calibration frequency of ground station data, which reflect the reliability of the data, are calculated. The quality indicators are transformed into fusion weights, and candidate values ​​for the initial analysis field are generated through weighted fusion. The error distribution of each data source is simulated based on the Monte Carlo method, generating multiple sets of input samples with errors. These samples are then input into a three-dimensional variational assimilation module to obtain multiple analysis field results. The mean and variance of the analysis fields are statistically analyzed to quantify the prediction uncertainty of each grid point. For thunderstorm cells and low-level wind shear zones in low-altitude meteorology, the background error covariance matrix of the three-dimensional variational assimilation is adjusted to increase the vertical background error variance of wind field and humidity in the 0-5km altitude layer, reduce the smoothing effect of large-scale background field on small-scale features, and preserve the vertical structural details of the convective system.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the low-altitude meteorological multi-source data processing method as described in any one of claims 1 to 8.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the low-altitude meteorological multi-source data processing method as described in any one of claims 1 to 8.

Images