Meteorological prediction method, electronic device and readable storage medium

Abstract

The embodiment of the application provides a kind of meteorological prediction method, electronic equipment and readable storage medium, it is related to computer technical field.The embodiment of the application can be combined with the good fitting capacity of meteorological prediction model, the target meteorological data of target time is accurately calculated according to initial meteorological data, to improve the precision of meteorological prediction.In addition, since the embodiment of the application can calculate the target meteorological data of target time, and target time can be any time, therefore, the embodiment of the application can update target meteorological cloud chart with higher frequency, so that the electronic device can control to display the latest meteorological cloud chart, to improve the response speed of meteorological disaster warning, alarm.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a weather forecasting method, electronic device, and readable storage medium. Background Technology

[0002] Weather forecasting is of great significance for preventing meteorological disasters. Among related technologies, the commonly used weather forecasting method is to judge the trend of meteorological cloud maps by manual observation, so as to make weather forecasts.

[0003] However, relying on manual observation often leads to problems such as slow response, large judgment errors, and low accuracy. Therefore, how to effectively improve the accuracy and response speed of weather forecasts is an urgent problem to be solved. Summary of the Invention

[0004] In view of this, embodiments of this application provide a weather forecasting method, an electronic device, and a readable storage medium to improve the accuracy and update frequency of weather forecasts, thereby improving the response speed of weather disaster early warnings and alarms.

[0005] Firstly, a weather forecasting method is provided, the method comprising:

[0006] Determine initial meteorological data, which includes multiple meteorological pixels and meteorological data corresponding to each meteorological pixel.

[0007] The meteorological data of the target pixel in the meteorological pixels, and the associated data of the target pixel, are input into a pre-trained meteorological prediction model to determine the target meteorological data of the target pixel at the target time. The associated data is used to characterize the meteorological data that has a correlation with the target pixel in the time dimension and / or spatial dimension.

[0008] A target meteorological cloud map is generated based on the initial meteorological data and the target meteorological data.

[0009] The control displays the target weather cloud map.

[0010] In some embodiments, determining the initial meteorological data includes:

[0011] Acquire initial radar data, which includes meteorological data collected by multiple weather radars at different elevation angles.

[0012] The initial radar data is filtered and interpolated to determine the initial meteorological data.

[0013] In some embodiments, determining the initial meteorological data includes:

[0014] Acquire initial satellite data, which includes cloud image data collected by at least multiple satellites.

[0015] The cloud image data in the initial satellite data are fused to determine the initial meteorological data.

[0016] In some embodiments, the meteorological prediction model includes persistent precipitation prediction models and convective precipitation prediction models for multiple predetermined geographical areas.

[0017] The step of inputting the meteorological data of the target pixel in the meteorological pixels, and the associated data of the target pixel, into a pre-trained meteorological prediction model to determine the target meteorological data of the target pixel at the target time includes:

[0018] Based on the predetermined geographical region where the target pixel is located and the precipitation type corresponding to the target pixel, the target prediction model corresponding to the target pixel is determined in the meteorological prediction model.

[0019] The meteorological data of the target pixel and the associated data of the target pixel are input into the target prediction model to determine the target meteorological data of the target pixel at the target time.

[0020] In some embodiments, the method further includes:

[0021] Obtain topographic maps of the catchment area and diagrams of railway lines, bridges, and culverts.

[0022] An overlay map is generated based on the target meteorological cloud map, the catchment area topographic map, and the route bridge and culvert equipment map.

[0023] Control the display of the overlay map.

[0024] In some embodiments, the method further includes:

[0025] Based on the target meteorological cloud map and the pre-trained strong convective cloud cluster tracking and prediction model, the predicted rainfall location and predicted rainfall amount corresponding to the target meteorological cloud map are determined.

[0026] The control displays the predicted rainfall location and the predicted rainfall amount.

[0027] In some embodiments, the control of displaying the predicted rainfall location and the predicted rainfall amount includes:

[0028] Obtain a railway network map, which includes the target railway route.

[0029] Based on the railway network map, the predicted rainfall location, and the predicted rainfall amount, the system controls the display of the predicted rainfall location and predicted rainfall amount within a predetermined range around the target railway route.

[0030] In some embodiments, the method further includes:

[0031] A meteorological disaster warning is issued in response to the predicted rainfall exceeding a predetermined rainfall threshold within a predetermined range around the target railway line.

[0032] In response to the real-time rainfall exceeding a predetermined rainfall threshold within a predetermined range around the target railway line, a meteorological disaster alarm is triggered, wherein the real-time rainfall is determined at least based on a rain gauge.

[0033] Secondly, a weather forecasting device is provided, the device comprising:

[0034] The initial meteorological data determination module is configured to determine initial meteorological data, which includes multiple meteorological pixels and meteorological data corresponding to each meteorological pixel.

[0035] The meteorological data prediction module is configured to input the meteorological data of the target pixel in the meteorological pixels and the associated data of the target pixel into a pre-trained meteorological prediction model to determine the target meteorological data of the target pixel at a target time, wherein the associated data is used to characterize the meteorological data that has a correlation with the target pixel in the time dimension and / or spatial dimension.

[0036] The meteorological cloud map generation module is configured to generate a target meteorological cloud map based on the initial meteorological data and the target meteorological data.

[0037] The first display module is configured to control and display the target meteorological cloud map.

[0038] In some embodiments, the initial meteorological data determination module is specifically configured to perform:

[0039] Acquire initial radar data, which includes meteorological data collected by multiple weather radars at different elevation angles.

[0040] The initial radar data is filtered and interpolated to determine the initial meteorological data.

[0041] In some embodiments, the initial meteorological data determination module is specifically configured to perform:

[0042] Acquire initial satellite data, which includes cloud image data collected by at least multiple satellites.

[0043] The cloud image data in the initial satellite data are fused to determine the initial meteorological data.

[0044] In some embodiments, the meteorological prediction model includes persistent precipitation prediction models and convective precipitation prediction models for multiple predetermined geographical areas.

[0045] The meteorological data prediction module is specifically configured to execute:

[0046] Based on the predetermined geographical region where the target pixel is located and the precipitation type corresponding to the target pixel, the target prediction model corresponding to the target pixel is determined in the meteorological prediction model.

[0047] The meteorological data of the target pixel and the associated data of the target pixel are input into the target prediction model to determine the target meteorological data of the target pixel at the target time.

[0048] In some embodiments, the apparatus further includes:

[0049] The acquisition module is configured to acquire topographic maps of catchment areas and diagrams of railway lines, bridges, and culverts.

[0050] The overlay map generation module is configured to generate an overlay map based on the target meteorological cloud map, the catchment area topographic map, and the route bridge and culvert equipment map.

[0051] The second display module is configured to control the display of the overlay map.

[0052] In some embodiments, the apparatus further includes:

[0053] The rainfall prediction module is configured to determine the predicted rainfall location and predicted rainfall amount corresponding to the target meteorological cloud map based on the target meteorological cloud map and a pre-trained strong convective cloud cluster tracking and prediction model.

[0054] The third display module is configured to control and display the predicted rainfall location and the predicted rainfall amount.

[0055] In some embodiments, the third display module is specifically configured to perform:

[0056] Obtain a railway network map, which includes the target railway route.

[0057] Based on the railway network map, the predicted rainfall location, and the predicted rainfall amount, the system controls the display of the predicted rainfall location and predicted rainfall amount within a predetermined range around the target railway route.

[0058] In some embodiments, the apparatus further includes:

[0059] The early warning module is configured to execute a meteorological disaster early warning in response to a predicted rainfall amount exceeding a predetermined rainfall threshold within a predetermined range around the target railway line.

[0060] The alarm module is configured to execute a meteorological disaster alarm in response to a predetermined rainfall amount within a predetermined range around the target railway line exceeding a predetermined rainfall threshold, wherein the real-time rainfall amount is determined at least based on a rain gauge.

[0061] Thirdly, embodiments of this application provide an electronic device, including a memory and a processor, wherein the memory is used to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method as described in the first aspect.

[0062] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer program instructions thereon, which, when executed by a processor, implement the method described in the first aspect.

[0063] In this embodiment, initial meteorological data can be determined, and the meteorological data of the target pixel and its associated data are input into a pre-trained meteorological prediction model to determine the target meteorological data of the target pixel at the target time. Furthermore, this embodiment can generate a target meteorological cloud map and control its display based on the initial and target meteorological data. In this process, combined with the good fitting ability of the meteorological prediction model, this embodiment can accurately calculate the target meteorological data at the target time based on the initial meteorological data, thereby improving the accuracy of meteorological prediction. In addition, since this embodiment can calculate the target meteorological data at the target time, and the target time can be any time, this embodiment can update the target meteorological cloud map at a higher frequency, enabling the electronic device to control the display of the latest meteorological cloud map, thereby improving the response speed of meteorological disaster warnings and alarms. Attached Figure Description

[0064] The above and other objects, features and advantages of the present application will become clearer from the following description of embodiments of the present application with reference to the accompanying drawings, in which:

[0065] Figure 1 This is a schematic diagram of the meteorological forecasting system in the embodiments of this application;

[0066] Figure 2 This is a schematic diagram of another weather forecasting system in the embodiments of this application;

[0067] Figure 3 This is a flowchart of the meteorological forecasting method in the embodiments of this application;

[0068] Figure 4 This is a schematic diagram of radar echo data in an embodiment of this application;

[0069] Figure 5 This is a schematic diagram of the display interface in an embodiment of this application;

[0070] Figure 6 This is a schematic diagram of the structure of the weather forecasting device in the embodiments of this application;

[0071] Figure 7 This is a schematic diagram of the structure of the electronic device in the embodiments of this application. Detailed Implementation

[0072] The present application is described below based on embodiments, but it is not limited to these embodiments. In the detailed description of the present application below, certain specific details are described in detail. Those skilled in the art can fully understand the present application without these details. To avoid obscuring the substance of the present application, well-known methods, processes, flows, elements, and circuits are not described in detail.

[0073] Furthermore, those skilled in the art should understand that the accompanying drawings provided herein are for illustrative purposes only and are not necessarily drawn to scale.

[0074] Unless the context explicitly requires it, words such as "including" or "contains" in the instruction manual should be interpreted as including rather than exclusive or exhaustive; that is, meaning "including but not limited to".

[0075] In the description of this application, it should be understood that the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, in the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0076] In related technologies, a common method of weather forecasting is to manually observe and judge the trend of weather cloud images to make weather predictions. For example, observers can use meteorological data obtained from public sources to judge the trend of weather cloud images, thereby determining whether rainfall or other phenomena will occur at a certain location within a certain time period.

[0077] However, on the one hand, the update frequency of meteorological data from public channels is low (it is often updated only every few minutes), which means that the data observed by observers may be somewhat delayed, thus affecting the accuracy and response speed of weather forecasts to some extent. On the other hand, since manual observation cannot cover all areas, it also affects the accuracy and response speed of weather forecasts to some extent. Therefore, in related technologies, manual observation often leads to problems such as untimely response, large judgment errors, and low judgment accuracy.

[0078] To address the aforementioned problems, this application provides a weather forecasting method to effectively improve the accuracy and speed of weather forecasting. Specifically, this weather forecasting method can be applied to a weather forecasting system, which may include a data receiving unit, a data processing unit, and a display unit.

[0079] like Figure 1 As shown, Figure 1 This is a schematic diagram of a weather forecasting system in an embodiment of this application. The data receiving unit 11 can be configured to receive weather data and send it to the data processing unit 12. The data processing unit 12 can be a processor in an electronic device, which can be a terminal or a server. The terminal can be a smartphone, tablet, or personal computer (PC), etc. The server can be a single server, a distributed server cluster, or a cloud server. The data processing unit 12 can be configured to receive the weather data sent by the data receiving unit 11, execute the aforementioned weather forecasting method, and control the display unit 13. The display unit 13 can be configured to receive control commands sent by the data processing unit 12 and display the corresponding interface.

[0080] The data processing unit 12 may include a software system. The web system of this software system may be a system built on the Java architecture. On the mobile terminal, this software system may be a weather forecasting system APP. The mobile terminal interface service and the mobile terminal interface forwarding service may both adopt the Spring Boot architecture.

[0081] Specifically, such as Figure 2As shown, the logical architecture of the meteorological forecasting system 21 can be divided into five basic layers: infrastructure layer 211, data layer 212, technical support layer 213, business application layer 214, and user terminal layer 215. Infrastructure layer 211 can include network and hardware devices, operating system, and related system software. Data layer 212 can include a centralized database and file services to store structured, file-based data. Technical support layer 213 can include the Java-based development framework SpringMVC, the database framework Mybatis, the automated build tool Maven, the storage system Redis, and the WebGIS engine Cesium. Business application layer 214 can implement functions such as rainfall forecasting, disaster alarms, rain gauge alarms, four-color early warning, earthquake emergency response, and railway equipment management. User terminal layer 215 can be used for user browsing operations, including page layout, style, and integrated display of business data, completing user interaction.

[0082] In addition, the data in this weather forecasting system 21 can include structured and unstructured data. Structured data can be divided into master data, operational data, and analytical data. Master data can include common basic data such as GIS data, lines, bridges and culverts, ground stations, catchment areas, radar, and rain gauge data. Operational data can include all transactional and management data, such as rainfall data, meteorological data, earthquake data, debris flow data, employee data, and rain gauge data. Analytical data can include report data, statistical data, and historical data.

[0083] Furthermore, in Figure 1 In this context, the data receiving unit 11 can be a device in an electronic device used to receive meteorological data. It can receive meteorological data by communicating with meteorological data acquisition equipment outside the electronic device. Alternatively, the data receiving unit 11 can be a device in an electronic device directly used to collect meteorological data.

[0084] The display unit 13 can be a display screen or other display device installed in an electronic device, which can directly receive control commands sent by the data processing unit 12 and display the corresponding interface. The display unit 13 can also be an independent display device outside the electronic device, which can receive control commands generated by the data processing unit 12 and display the corresponding interface through a wired or wireless communication connection with the electronic device.

[0085] Furthermore, such as Figure 3 As shown, the data processing unit 12 can be configured to perform the following steps:

[0086] In step S110, initial meteorological data are determined.

[0087] The initial meteorological data can be obtained from public sources or preprocessed data obtained from those public sources. The initial meteorological data includes multiple meteorological pixels and the corresponding meteorological data for each pixel. In other words, the initial meteorological data can be displayed in the form of images such as meteorological cloud maps. Each pixel in the image can represent the meteorological data within its corresponding area, which may include precipitation, temperature, wind field, and other data.

[0088] In an optional implementation, step S110 may include the following steps:

[0089] In step S111, initial radar data is acquired.

[0090] The initial radar data includes meteorological data collected by meteorological radars at at least multiple elevation angles. In practical applications, typical meteorological radar images only have single-layer elevation angle data, which often results in negative effects such as super-refractional clutter and missing fan-shaped patterns, leading to low accuracy in weather forecasting. In this embodiment, by acquiring meteorological data collected by meteorological radars at multiple elevation angles, the problems caused by super-refractional clutter and missing fan-shaped patterns can be avoided.

[0091] In step S112, the initial radar data is filtered and interpolated to determine the initial meteorological data.

[0092] The filtering process can include three-dimensional filtering, continuous characteristic filtering, morphological filtering, and speckle noise filtering, while the interpolation process can include spatial interpolation extension methods.

[0093] like Figure 4 As shown, Figure 4 This is a schematic diagram of the echo data 41 corresponding to the initial meteorological data in this embodiment of the application.

[0094] Depend on Figure 4 It is known that traditional radar echo data contains negative interference information such as super-refractive clutter and sector missing. The filtering and interpolation processing in the embodiments of this application can effectively filter out the above-mentioned negative interference information, thereby obtaining echo data that is more consistent with the actual situation.

[0095] In this embodiment, since the initial radar data includes meteorological data collected by meteorological radars at at least multiple elevation angles, this embodiment can perform filtering processes such as three-dimensional filtering, continuity characteristic filtering, morphological filtering, and speckle noise filtering on the meteorological data at multiple elevation angles to remove non-precipitation clutter such as clear-sky ground object echoes and insect / bird flock echoes. Furthermore, this embodiment can also employ a multi-layer elevation angle echo spatial interpolation extension method to compensate for missing radial data in the meteorological radar image, ensuring sufficient quality for both real-time and extrapolated radar images.

[0096] In an optional implementation, step S110 may also include the following steps:

[0097] In step S113, initial satellite data is acquired.

[0098] The initial satellite cloud image data includes cloud image data collected by at least multiple satellites. In practical applications, the means of acquiring meteorological data for certain geographical areas are limited (for example, there may be insufficient numbers of meteorological radars in certain geographical areas). In this case, the embodiments of this application can determine the initial meteorological data by using meteorological satellites with a large coverage area.

[0099] In addition, the satellites used to collect cloud image data in this application embodiment can be Fengyun series satellites, Kuihua series satellites, MTG meteorological satellites, Geostationary Operational Environmental Satellites (GOES), etc.

[0100] In step S114, the cloud image data in the initial satellite data are fused to determine the initial meteorological data.

[0101] Through the embodiments of this application, cloud image data collected by multiple satellites can be used as initial satellite data. Then, the cloud image data in the initial satellite data are fused to determine the initial meteorological data. In this way, the means of determining initial meteorological data can be expanded, and the adaptability of the meteorological forecasting system can be improved.

[0102] In addition, in conjunction with the above-described embodiments, the embodiments of this application can also simultaneously acquire the above-described initial radar data and the above-described initial satellite data, and then further fuse the two types of data to determine the initial meteorological data, so as to make the initial meteorological data more accurate.

[0103] In step S120, the meteorological data of the target pixel in the meteorological pixel and the associated data of the target pixel are input into the pre-trained meteorological prediction model to determine the target meteorological data of the target pixel at the target time.

[0104] The associated data is used to characterize meteorological data that has a correlation with the target pixel in the time and / or spatial dimensions. The meteorological prediction model can be a machine learning model (e.g., a neural network model built based on a convolutional neural network). In this embodiment, the machine learning model can be trained based on historical data to obtain the meteorological prediction model. The historical data can include historical time points, radar images at a certain historical moment, short-term rainfall images at a certain historical moment, warning catchment areas, and related attribute information of the catchment areas, etc.

[0105] In this embodiment, the meteorological prediction model trained on historical data can predict weather changes based on the learned historical data. When encountering meteorological disasters (such as floods, rainstorms, etc.), the meteorological prediction model can realistically reflect the situation at the time of the meteorological disaster based on historical data from past disasters. For some major meteorological disaster events, such as the series of impacts of meteorological disasters on traffic safety, roadbeds and bridges, and personnel safety, this embodiment can train the meteorological prediction model by retrospectively analyzing major events, thereby establishing an effective prevention mechanism to prevent similar major accidents from recurring.

[0106] In one alternative implementation, the meteorological forecasting model may include persistent precipitation forecasting models and convective precipitation forecasting models for multiple predetermined geographical areas.

[0107] Because different geographical regions have different climate characteristics and altitudes, even if two different geographical regions have the same initial meteorological data, they may still have different climate characteristics. Taking meteorological radar data as an example, due to the different climate characteristics and altitudes of different geographical regions, the radar echo intensity of the same rainfall may have significant differences. Therefore, the embodiments of this application can train meteorological prediction models separately for different predetermined geographical regions, making each meteorological prediction model more targeted. The predetermined geographical regions can be divided according to terrain, latitude and longitude, or other applicable methods.

[0108] Furthermore, even within the same geographical region, persistent precipitation and convective precipitation exhibit different climatic characteristics. Therefore, in this embodiment of the application, persistent precipitation prediction models and convective precipitation prediction models can be set separately within the same predetermined geographical region to achieve separate predictions for persistent precipitation and convective precipitation, thereby further improving the accuracy of meteorological forecasts.

[0109] Furthermore, step S120 above may include the following steps:

[0110] In step S121, based on the predetermined geographical region where the target pixel is located and the precipitation type corresponding to the target pixel, the target prediction model corresponding to the target pixel is determined in the meteorological prediction model.

[0111] In this embodiment, since the meteorological prediction models are divided according to geographical regions and precipitation types, different meteorological prediction models can be used for different precipitation types in each predetermined geographical region, thereby improving the accuracy of meteorological prediction.

[0112] Furthermore, since time also has a certain impact on climate, such as the influence of seasons and different times of the same day, this application embodiment, based on the division of the meteorological prediction model according to geographical region and precipitation type, can further divide the meteorological prediction model in the time dimension to determine the meteorological prediction model corresponding to different time periods in each predetermined geographical region and precipitation type. The time period can be in units such as hours, days, weeks, and months.

[0113] In step S122, the meteorological data of the target pixel and the associated data of the target pixel are input into the target prediction model to determine the target meteorological data of the target pixel at the target time.

[0114] Through the embodiments of this application, target meteorological data for a target pixel at a target time can be determined based on initial meteorological data and the corresponding target prediction model. This enables the use of different meteorological prediction models for different precipitation types within various predetermined geographical areas, thereby improving the accuracy of meteorological forecasts.

[0115] In step S130, a target meteorological cloud map is generated based on the initial meteorological data and the target meteorological data.

[0116] The target meteorological cloud map may include initial meteorological data and target meteorological data for each target time. Specifically, embodiments of this application can determine the meteorological cloud map corresponding to the initial meteorological data and the meteorological cloud map corresponding to each target time, and then merge the various meteorological cloud maps to determine the target meteorological cloud map. The target meteorological cloud map may be a video-based meteorological cloud map, a dynamic image (e.g., a GIF-formatted dynamic image), or a collection of static images.

[0117] In step S140, the target weather cloud map is displayed.

[0118] In this embodiment of the application, the electronic device used to execute the meteorological forecasting method can control its own or an externally connected display screen to display the target meteorological cloud map after generating the target meteorological cloud map through control commands.

[0119] In this embodiment, initial meteorological data can be determined, and the meteorological data of the target pixel and its associated data are input into a pre-trained meteorological prediction model to determine the target meteorological data of the target pixel at the target time. Furthermore, this embodiment can generate a target meteorological cloud map and control its display based on the initial and target meteorological data. In this process, combined with the good fitting ability of the meteorological prediction model, this embodiment can accurately calculate the target meteorological data at the target time based on the initial meteorological data, thereby improving the accuracy of meteorological prediction. In addition, since this embodiment can calculate the target meteorological data at the target time, and the target time can be any time, this embodiment can update the target meteorological cloud map at a higher frequency, enabling the electronic device to control the display of the latest meteorological cloud map, thereby improving the response speed of meteorological disaster warnings and alarms.

[0120] In one optional implementation, the embodiments of this application may further include the following steps:

[0121] In step S210, a topographic map of the catchment area and a diagram of the railway line, bridges, and culverts are obtained.

[0122] Among them, the topographic map of the catchment area can be drawn based on the topographic data obtained from public channels, and the map of the railway line, bridge and culvert equipment can be drawn based on the data of the railway line, bridge and culvert equipment obtained from public channels.

[0123] In step S220, an overlay map is generated based on the target meteorological cloud map, the catchment area topographic map, and the route bridge and culvert equipment diagram.

[0124] In this embodiment of the application, the target meteorological cloud map can represent the weather conditions, the catchment area topographic map and the line bridge and culvert equipment map can represent the landform and the line bridge and culvert equipment on the ground. Therefore, by superimposing the above three images, this embodiment of the application can determine the weather conditions near each geographical location and each line bridge and culvert equipment, thereby knowing whether a meteorological disaster is about to occur or is occurring near each geographical location and each facility.

[0125] In step S230, the overlay map is displayed.

[0126] In this embodiment of the application, the electronic device used to perform the weather forecasting method can control its own or an externally connected display screen to display the overlay map after generating the overlay map through control commands.

[0127] This application embodiment, by controlling the display overlay map, can visually show the meteorological disasters that are about to occur or are occurring near various geographical locations and bridge and culvert equipment, thereby achieving the purpose of timely early warning and alarm.

[0128] In one optional implementation, the embodiments of this application may further include the following steps:

[0129] In step S310, based on the target meteorological cloud map and the pre-trained strong convective cloud cluster tracking and prediction model, the predicted rainfall location and predicted rainfall amount corresponding to the target meteorological cloud map are determined.

[0130] The strong convective cloud tracking and prediction model can be a machine learning model (e.g., a neural network model built based on a convolutional neural network). In this embodiment, the machine learning model can be trained based on historical data to obtain the strong convective cloud tracking and prediction model. The historical data can include historical time points, radar images at a certain historical moment, short-term rainfall images at a certain historical moment, warning catchment areas, and related attribute information of the catchment areas.

[0131] After determining an accurate target meteorological cloud map in the embodiments of this application, the rainfall situation at each geographical location can be accurately determined based on the pre-trained strong convective cloud cluster tracking and prediction model and the accurate target meteorological cloud map.

[0132] In step S320, the system displays the predicted rainfall location and predicted rainfall amount.

[0133] In this embodiment of the application, the electronic device used to perform the meteorological forecasting method can, after determining the predicted rainfall location and the predicted rainfall amount, control its own or an externally connected display screen to display the predicted rainfall location and the predicted rainfall amount through control commands.

[0134] This application embodiment, by controlling the display of predicted rainfall location and predicted rainfall amount, can visually display the rainfall situation that is about to occur or is occurring near various geographical locations, thereby achieving the purpose of timely early warning and alarm.

[0135] In an optional implementation, step S320 may include the following steps:

[0136] In step S321, a railway network map is obtained, which includes the target railway route.

[0137] The railway network map is railway network data obtained from public channels in this application embodiment. The target railway route can be all railway routes in the above-mentioned railway network map, or it can be a part of the railway routes in the above-mentioned railway network map.

[0138] In step S322, based on the railway network map, predicted rainfall location, and predicted rainfall amount, the system controls the display of predicted rainfall location and predicted rainfall amount within a predetermined range around the target railway line.

[0139] The predetermined range around the target railway line can be any applicable value such as 500 meters, 1000 meters, or 2000 meters around the railway line.

[0140] like Figure 5 As shown, Figure 5 This is a schematic diagram of a display interface in an embodiment of this application.

[0141] Depend on Figure 5 It can be seen that, Figure 5 The display interface shown includes a railway network map, predicted rainfall location, and predicted rainfall amount. By predicting the rainfall location and predicted rainfall amount, this embodiment of the application can accurately predict the rainfall situation near each railway line.

[0142] Since railway flood prevention is of paramount importance in ensuring railway safety, and these disasters are often related to rainfall, this application embodiment uses a railway network map, predicted rainfall locations, and predicted rainfall amounts to display the predicted rainfall locations and amounts within a predetermined range around the target railway line, thereby enabling rapid and timely railway flood prevention early warning and alarm.

[0143] In addition, with Figure 5 Taking the displayed interface as an example, this interface can provide a global radar extrapolation image for the next 2 hours, with a dynamic playback effect of radar trends via a timeline. In practical applications, staff can click on any point to query precipitation forecasts for the next 2 hours and 24 hours in 5-minute increments. This interface can also provide radar extrapolation images of any station for the next 2 hours, with a dynamic playback effect of radar trends via a timeline. Staff can click on any point to query precipitation forecasts for the next 2 hours in 5-minute increments and 24 hours in 1-hour increments. This interface can also provide global precipitation forecast trend maps for the next 24 hours, 72 hours, and 15 days. Staff can click on any point to query precipitation forecasts for the next 24 hours, 72 hours in 1-hour increments, and 15 days in 1-day increments. This interface can also provide a global temperature trend color-coded map for the previous 48 hours and a curve change map for the next 24 hours, updated hourly to allow viewing of historical temperatures and predicting future temperatures. This interface can also provide a global humidity trend color-coded map for the previous 48 hours and a curve change map for the next 24 hours, updated hourly to allow viewing of historical humidity and predicting future humidity. The display interface also provides a color-coded map of global wind trends for the past 48 hours and a curve showing changes for the next 24 hours, updating hourly to allow for viewing historical wind speeds and predicting future wind speeds. Furthermore, the interface allows users to immediately query information such as magnitude, epicenter, and affected area after an earthquake, and provides alerts.

[0144] In other words, this application embodiment, while predicting the location and amount of rainfall, can also predict weather conditions such as temperature, humidity, and wind force based on initial meteorological data and corresponding machine learning models, thereby making the information displayed by the electronic device more diversified. Furthermore, this application embodiment can also query information such as magnitude, epicenter, and affected area of ​​an earthquake immediately after it occurs based on publicly available data, and issue an alarm to achieve timely warnings of earthquake disasters.

[0145] It should be noted that the above Figure 5 This is just one example of an embodiment of the present application. In practical applications, the data update frequency in the above example can be set according to the actual situation. The present application does not limit the data update frequency.

[0146] In one optional implementation, the embodiments of this application may further include the following steps:

[0147] In step S410, a meteorological disaster warning is issued in response to the predicted rainfall exceeding a predetermined rainfall threshold within a predetermined range around the target railway line.

[0148] The predetermined rainfall threshold can be a single threshold or a tiered threshold composed of multiple thresholds. Each tier can correspond to different meteorological disaster warnings. For example, each tier can correspond to a blue warning, a yellow warning, an orange warning, and a red warning. Alternatively, each tier can correspond to a level one warning, a level two warning, a level three warning, and a level four warning. The embodiments of this application can set the warning names corresponding to each tier according to actual circumstances.

[0149] In step S420, a meteorological disaster alarm is triggered in response to the real-time rainfall exceeding a predetermined rainfall threshold within a predetermined range around the target railway line.

[0150] The real-time rainfall is determined at least based on a rain gauge. The predetermined rainfall threshold can be a single threshold or a tiered threshold composed of multiple thresholds. Each tier can correspond to different meteorological disaster alarms; for example, each tier can correspond to a blue alarm, a yellow alarm, an orange alarm, and a red alarm. Alternatively, each tier can correspond to a level 1 alarm, a level 2 alarm, a level 3 alarm, and a level 4 alarm. In this embodiment, the alarm name corresponding to each tier can be set according to actual conditions.

[0151] Based on the same technical concept, embodiments of this application also provide a weather forecasting device, such as... Figure 6 As shown, the device includes: an initial meteorological data determination module 61, a meteorological data prediction module 62, a meteorological cloud map generation module 63, and a display module 64.

[0152] The initial meteorological data determination module 61 is configured to determine initial meteorological data, which includes multiple meteorological pixels and meteorological data corresponding to each meteorological pixel.

[0153] The meteorological data prediction module 62 is configured to input the meteorological data of the target pixel in the meteorological pixels and the associated data of the target pixel into a pre-trained meteorological prediction model to determine the target meteorological data of the target pixel at a target time, wherein the associated data is used to characterize the meteorological data that has a correlation with the target pixel in the time dimension and / or spatial dimension.

[0154] The meteorological cloud map generation module 63 is configured to generate a target meteorological cloud map based on the initial meteorological data and the target meteorological data.

[0155] The first display module 64 is configured to control and display the target meteorological cloud map.

[0156] In some embodiments, the initial meteorological data determination module 61 is specifically configured to perform:

[0157] Acquire initial radar data, which includes meteorological data collected by multiple weather radars at different elevation angles.

[0158] The initial radar data is filtered and interpolated to determine the initial meteorological data.

[0159] In some embodiments, the initial meteorological data determination module 61 is specifically configured to perform:

[0160] Acquire initial satellite data, which includes cloud image data collected by at least multiple satellites.

[0161] The cloud image data in the initial satellite data are fused to determine the initial meteorological data.

[0162] In some embodiments, the meteorological prediction model includes persistent precipitation prediction models and convective precipitation prediction models for multiple predetermined geographical areas.

[0163] The meteorological data prediction module 62 is specifically configured to execute:

[0164] Based on the predetermined geographical region where the target pixel is located and the precipitation type corresponding to the target pixel, the target prediction model corresponding to the target pixel is determined in the meteorological prediction model.

[0165] The meteorological data of the target pixel and the associated data of the target pixel are input into the target prediction model to determine the target meteorological data of the target pixel at the target time.

[0166] In some embodiments, the apparatus further includes:

[0167] The acquisition module is configured to acquire topographic maps of catchment areas and diagrams of railway lines, bridges, and culverts.

[0168] The overlay map generation module is configured to generate an overlay map based on the target meteorological cloud map, the catchment area topographic map, and the route bridge and culvert equipment map.

[0169] The second display module is configured to control the display of the overlay map.

[0170] In some embodiments, the apparatus further includes:

[0171] The rainfall prediction module is configured to determine the predicted rainfall location and predicted rainfall amount corresponding to the target meteorological cloud map based on the target meteorological cloud map and a pre-trained strong convective cloud cluster tracking and prediction model.

[0172] The third display module is configured to control and display the predicted rainfall location and the predicted rainfall amount.

[0173] In some embodiments, the third display module is specifically configured to perform:

[0174] Obtain a railway network map, which includes the target railway route.

[0175] Based on the railway network map, the predicted rainfall location, and the predicted rainfall amount, the system controls the display of the predicted rainfall location and predicted rainfall amount within a predetermined range around the target railway route.

[0176] In some embodiments, the apparatus further includes:

[0177] The early warning module is configured to execute a meteorological disaster early warning in response to a predicted rainfall amount exceeding a predetermined rainfall threshold within a predetermined range around the target railway line.

[0178] The alarm module is configured to execute a meteorological disaster alarm in response to a predetermined rainfall amount within a predetermined range around the target railway line exceeding a predetermined rainfall threshold, wherein the real-time rainfall amount is determined at least based on a rain gauge.

[0179] In this embodiment, initial meteorological data can be determined, and the meteorological data of the target pixel and its associated data are input into a pre-trained meteorological prediction model to determine the target meteorological data of the target pixel at the target time. Furthermore, this embodiment can generate a target meteorological cloud map and control its display based on the initial and target meteorological data. In this process, combined with the good fitting ability of the meteorological prediction model, this embodiment can accurately calculate the target meteorological data at the target time based on the initial meteorological data, thereby improving the accuracy of meteorological prediction. In addition, since this embodiment can calculate the target meteorological data at the target time, and the target time can be any time, this embodiment can update the target meteorological cloud map at a higher frequency, enabling the electronic device to control the display of the latest meteorological cloud map, thereby improving the response speed of meteorological disaster warnings and alarms.

[0180] Figure 7 This is a schematic diagram of an electronic device according to an embodiment of this application. For example... Figure 7 As shown, Figure 7 The illustrated electronic device is a general address lookup device, comprising a general computer hardware architecture, including at least a processor 71 and a memory 72. The processor 71 and memory 72 are connected via a bus 73. The memory 72 is adapted to store instructions or programs executable by the processor 71. The processor 71 can be a standalone microprocessor or a collection of one or more microprocessors. Thus, the processor 71 executes the instructions stored in the memory 72 to perform the method flow described in the embodiments of this application, thereby realizing data processing and control of other devices. The bus 73 connects the aforementioned components together, and also connects these components to a display controller 74, a display device, and an input / output (I / O) device 75. The input / output (I / O) device 75 can be a mouse, keyboard, modem, network interface, touch input device, motion-sensing input device, printer, and other devices known in the art. Typically, the input / output device 75 is connected to the system via an input / output (I / O) controller 76.

[0181] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus (devices), or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0182] This application is described with reference to flowchart illustrations of methods, apparatus (devices), and computer program products according to embodiments of this application. It should be understood that each step in the flowchart can be implemented by computer program instructions.

[0183] These computer program instructions may be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including an instruction means, the implementation process of which is described in the instruction means. Figure 1 The function specified in one or more processes.

[0184] These computer program instructions may also be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing device, produce instructions for implementing processes. Figure 1 A device for a function specified in one or more processes.

[0185] Another embodiment of this application relates to a non-volatile storage medium for storing a computer-readable program for use by a computer to execute some or all of the above-described method embodiments.

[0186] That is, those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program specifying the relevant hardware. This program is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods described in the embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0187] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A weather prediction method characterized by, The method comprises: determining initial meteorological data, the initial meteorological data comprising a plurality of meteorological pixels and meteorological data corresponding to each of the meteorological pixels; inputting meteorological data of a target pixel among the meteorological pixels and associated data of the target pixel into a pre-trained meteorological prediction model to determine target meteorological data of the target pixel at a target time, wherein the associated data is used to represent meteorological data having a correlation relationship with the target pixel in a time dimension and / or a space dimension, and the meteorological prediction model comprises a plurality of persistent precipitation prediction models and convective precipitation prediction models in predetermined geographical regions; generating a target meteorological cloud image according to the initial meteorological data and the target meteorological data; and controlling display of the target meteorological cloud image. The inputting of the meteorological data of the target pixel among the meteorological pixels and the associated data of the target pixel into the pre-trained meteorological prediction model to determine the target meteorological data of the target pixel at the target time comprises: dividing each meteorological prediction model according to geographical regions and precipitation types to use different meteorological prediction models for different precipitation types in each predetermined geographical region; dividing the meteorological prediction model according to a time dimension to determine meteorological prediction models corresponding to different time periods in each predetermined geographical region and each precipitation type, the time periods being in units of hours, days, weeks, and months; determining a target prediction model corresponding to the target pixel in the meteorological prediction model according to a predetermined geographical region where the target pixel is located and a precipitation type corresponding to the target pixel; inputting the meteorological data of the target pixel and the associated data of the target pixel into the target prediction model to determine the target meteorological data of the target pixel at the target time.

2. The method of claim 1, wherein, The determination of the initial meteorological data comprises: obtaining initial radar data, the initial radar data comprising meteorological data collected by a plurality of meteorological radars at different elevations; and performing filtering processing and interpolation processing on the initial radar data to determine the initial meteorological data.

3. The method of claim 1, wherein, The determination of the initial meteorological data comprises: obtaining initial satellite cloud image data, the initial satellite cloud image data comprising cloud image data collected by a plurality of satellites; and performing fusion processing on each cloud image data in the initial satellite cloud image data to determine the initial meteorological data.

4. The method of claim 1, wherein, The method further comprises: obtaining a catchment area topographic map and a line bridge culvert equipment map; generating an overlay map according to the target meteorological cloud image, the catchment area topographic map, and the line bridge culvert equipment map; and controlling display of the overlay map.

5. The method of claim 1, wherein, The method further comprises: determining a predicted rainfall position and a predicted rainfall amount corresponding to the target meteorological cloud image according to the target meteorological cloud image and a pre-trained severe convective cloud cluster tracking prediction model; and controlling display of the predicted rainfall position and the predicted rainfall amount.

6. The method of claim 5, wherein, The controlling of the display of the predicted rainfall position and the predicted rainfall amount comprises: obtaining a railway network map, the railway network map comprising a target railway line; and According to the railway network map, the predicted rainfall position and the predicted rainfall amount, a display is controlled to show the predicted rainfall position and the predicted rainfall amount within a predetermined range around the target railway route.

7. The method of claim 6, wherein, The method further comprises: in response to the predicted rainfall amount within a predetermined range around the target railway route being greater than a predetermined rainfall threshold, performing a meteorological disaster early warning; and in response to real-time rainfall amount within a predetermined range around the target railway route being greater than the predetermined rainfall threshold, performing a meteorological disaster alarm, the real-time rainfall amount being determined based on at least a pluviometer.

8. An electronic device comprising a memory and a processor, characterized in that The memory is configured to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method according to any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium has stored therein a computer program, which, when executed by a processor, implements the method according to any one of claims 1-7.

Images