A short-impending heavy rainfall forecasting method and system

Abstract

The application discloses a kind of short-acting forecast method and system of heavy rainfall in meteorological service technical field, comprising the following steps: S1: the monitored precipitation area is divided into first monitoring area;S2: meteorological station is as center in first area and carries out second area division;S3: based on the precipitation data acquisition of adjacent second monitoring area in meteorological detection station in adjacent second monitoring area, the precipitation change trend of adjacent second monitoring area is obtained;S4: based on the precipitation change trend of adjacent second monitoring area, adjust the position of second monitoring area, to find the position of precipitation center area;S5: based on the position of precipitation center area, extract the relevant characteristics of precipitation center area;S6: based on the trained machine learning model receives the relevant characteristics of precipitation center area and outputs heavy rainfall short-acting forecast result.The present application can determine the precipitation center area, combined with the existing forecast method based on machine learning, to realize a kind of short-acting forecast method and system of heavy rainfall.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of meteorological services, and specifically relates to a heavy rainfall short-impending forecast method and system. BACKGROUND

[0002] Heavy rainfall is a kind of severe convective weather phenomenon with high rainfall intensity in a short time, mainly occurring in the flood season every year, and often accompanied by strong wind, hail and thunderstorm and other severe weather phenomena, which is easy to cause natural disasters such as mountain torrents, mudslides and urban floods. The heavy rainfall short-impending forecast generally refers to the forecast of the heavy rainfall weather in a region in the next 0-12 hours. Improving the accuracy of the heavy rainfall short-impending forecast is one of the key research targets of meteorological researchers, but due to the high suddenness of heavy rainfall and the high-dimensional nonlinearity of meteorological data, the traditional short-impending forecast method has poor effect. Moreover, the traditional short-impending forecast method often only uses single meteorological data, and the model has poor anti-interference performance and is highly dependent on the accuracy of the data.

[0003] Under normal circumstances, the near-impending rainfall forecast for 0-2 hours is mainly based on radar or satellite observation data for extrapolation, while the short-impending rainfall forecast for 2-12 hours more relies on numerical weather prediction. According to the prediction method, the short-impending rainfall forecast is generally divided into statistical prediction method, numerical mode prediction method and extrapolation-based prediction method.

[0004] At present, radar data extrapolation is one of the main methods for short-impending rainfall forecast, but the traditional extrapolation method does not fully exploit the rules of historical observation data in terms of data utilization, and only relies on limited radar data for prediction. The precipitation area of the traditional radar echo extrapolation-based prediction is fuzzy and distorted, and there is a large error in the calculation of the precipitation amount according to the radar echo, so the position and intensity of the precipitation center are not very accurate. Although the ground meteorological automatic stations in China are distributed densely, it is still difficult for the ground automatic stations to capture the accurate position and intensity of the rainfall center due to the strong local characteristics of rainfall. Therefore, improving the accuracy of the observation of the position and intensity of the heavy rainfall center through technical methods to provide more accurate observation data for short-impending forecast can effectively improve the accuracy of the center value and position of the heavy rainfall forecast and determine the area with the maximum precipitation, so as to provide heavy rainfall warning for the public and industry departments and avoid potential safety risks. For the areas where mountain torrents and geological disasters frequently occur, accurate precipitation center forecast can help to take preventive measures in advance and reduce the loss caused by disasters. Therefore, it is urgent to propose a heavy rainfall short-impending forecast method and design a related system to improve the accuracy of the heavy rainfall short-impending forecast. SUMMARY

[0005] To solve the above problems, the application provides a heavy rainfall short-impending forecast method and system, which improves the accuracy and pertinence of the heavy rainfall short-impending forecast by finding the accurate position and intensity of the precipitation center.

[0006] To achieve the above object, the technical scheme of the present application is as follows:

[0007] In one aspect, a heavy rainfall short-impending forecast method is provided, comprising the following steps:

[0008] S1: monitoring a rainfall area, and dividing the rainfall area into a first monitoring area after the rainfall area is monitored;

[0009] S2: dividing the first monitoring area with a meteorological monitoring station as the center to obtain at least three second monitoring areas of regular shape, which are all located in the first monitoring area and have an area smaller than that of the first monitoring area;

[0010] S3: calculating the rainfall intensity distribution of different positions in the second monitoring area to which the meteorological monitoring station belongs and the total rainfall in the second monitoring area through the rainfall data of the same time period monitored by each meteorological monitoring station;

[0011] S4: judging the rainfall change trend in the second monitoring area and the rainfall change trend between adjacent second monitoring areas based on the rainfall intensity distribution and the total rainfall of each specified period corresponding to each second monitoring area;

[0012] S5: adjusting the position of the second monitoring area based on the rainfall change trend between adjacent second monitoring areas, and obtaining an overlapping area of the second monitoring areas; when the rainfall increase path directions of each second monitoring area are all towards the overlapping area of the second monitoring areas, extracting the overlapping area as a rainfall center area;

[0013] S6: extracting the rainfall-related features of the rainfall center area based on the position of the rainfall center area;

[0014] S7: receiving the rainfall-related features of the rainfall center area based on a trained machine learning model, and outputting a heavy rainfall short-impending forecast result.

[0015] Further, in S1, the rainfall area is divided into the first monitoring area, which comprises: determining the monitoring purpose, collecting basic data, analyzing the rainfall characteristics, and determining the monitoring area range.

[0016] Further, in S2, the first monitoring area is divided with the meteorological monitoring station as the center to obtain at least three second monitoring areas of regular shape, which comprises:

[0017] S201: obtaining the number and position of meteorological monitoring stations in the first monitoring area;

[0018] S202: determining the shape of the second monitoring area based on the range of the first monitoring area to ensure that the shape distribution can cover the first monitoring area;

[0019] S203: dividing the second monitoring area in a regular shape with the meteorological monitoring station as the center based on the number and position of the meteorological monitoring stations in the first monitoring area and determining the position of the second monitoring area;

[0020] S204: determining the size of the second monitoring area based on the shape and size of the second monitoring area.

[0021] Further, S3: calculating the precipitation intensity distribution at different positions in the second monitoring area to which the meteorological monitoring station belongs and the total precipitation in the second precipitation area based on the precipitation data of the same time period monitored by each meteorological monitoring station, including:

[0022] S301: performing radar detection by the meteorological monitoring stations in each second monitoring area based on the range of the second monitoring area;

[0023] S302: obtaining the precipitation intensity and precipitation data of the second monitoring area based on the radar detection by the meteorological monitoring stations in each second monitoring area;

[0024] S303: analyzing the precipitation intensity and precipitation data of each second monitoring area to obtain the precipitation intensity distribution in each second monitoring area and the total precipitation in the second precipitation area.

[0025] Further, analyzing the precipitation data of each second monitoring area to determine the trend of precipitation change, including:

[0026] By analyzing the precipitation data in the second monitoring area, the trend of precipitation change in the second monitoring area is obtained; by comparing the precipitation data of adjacent second monitoring areas, the size of the group of precipitation data is determined, and the trend of precipitation change in the precipitation change area is obtained.

[0027] Further, S5: adjusting the position of the second monitoring area based on the trend of precipitation change between adjacent second monitoring areas, and obtaining the overlapping area of the second monitoring area; when the directions of the precipitation increase paths of each second monitoring area are all towards the overlapping area of the second monitoring area, the overlapping area is extracted as the precipitation center area, including:

[0028] Based on the comparison of the trends of precipitation change in adjacent second monitoring areas, the paths of precipitation increase or decrease in the adjacent second monitoring areas are obtained. Through the analysis of the precipitation data, the second monitoring area in the precipitation change area will move in the direction of the precipitation increase path to the area with higher precipitation data. Based on the movement of each second monitoring area, the overlapping area of each second monitoring area is obtained; when the directions of the precipitation increase paths of each second monitoring area are all towards the overlapping area of the second monitoring area, the overlapping area is extracted as the precipitation center area.

[0029] Further, in S5, the module for adjusting the position of the second monitoring area based on the precipitation change trend between adjacent second monitoring areas to find the position of the precipitation center area, comprises:

[0030] S501: Collecting precipitation data of corresponding weather stations in each second monitoring area;

[0031] S502: Collecting and summarizing the precipitation data of weather stations in the second monitoring area within the first monitoring area;

[0032] S503: Comparing the precipitation data of the weather stations, judging the precipitation trend and the path of precipitation increase or decrease;

[0033] S504: Judging whether the edge of the second monitoring area is on the path of precipitation increase, and changing the size and position of each second monitoring area to obtain the overlapping area of each second monitoring area;

[0034] S505: Judging whether the direction of precipitation increase of each second monitoring area is towards the overlapping area of the second monitoring area, if not, returning to the precipitation data comparison unit, comparing the precipitation data in the direction of precipitation increase, outputting a larger path of precipitation increase, and adjusting the second monitoring area, until the direction of precipitation increase of each second monitoring area is towards the overlapping area of the second monitoring area, and extracting the overlapping area as the precipitation center area.

[0035] Further, in S6, based on the position of the precipitation center area, extracting the precipitation-related features of the precipitation center area, comprising: based on S1 to S5, determining the specific position of the precipitation center area, collecting the precipitation data of the precipitation center area, including historical data and real-time data; based on data analysis, the main factors affecting the precipitation characteristics of the precipitation center area; summarizing the main precipitation characteristics of the precipitation center area.

[0036] Further, in S7, the process of receiving the precipitation-related features of the precipitation center area based on the trained machine learning model to output the short-term heavy rain forecast result is: based on the main precipitation characteristics of the precipitation center area, inputting the features into the machine learning model which has been constructed and trained, then the machine learning model can predict the rainfall of the next moment of the precipitation center area, thereby obtaining the rainfall sequence of the future of the precipitation center area, and outputting the rainfall sequence can obtain the short-term heavy rain forecast data of the precipitation center area.

[0037] On the other hand, a short-term heavy rain forecasting system is provided, which is based on the short-term heavy rain forecasting method of any one of claims 1-9, characterized in that it comprises:

[0038] A monitoring module for monitoring a precipitation area, and when a precipitation area is monitored, dividing the precipitation area into a first monitoring area;

[0039] The division module is configured to divide the first monitoring area in a meteorological monitoring station as a center to obtain at least three second monitoring areas in regular shapes, and the second monitoring areas are located in the first monitoring area and have smaller areas than the first monitoring area;

[0040] The collection module is configured to calculate the rainfall intensity distribution of different positions in a second monitoring area to which the meteorological monitoring station belongs and the total rainfall in the second monitoring area by using the rainfall data of the same time period monitored by each meteorological monitoring station;

[0041] The judgment module is configured to judge the rainfall variation trend in each second monitoring area and the rainfall variation trend between adjacent second monitoring areas according to the rainfall intensity distribution and the total rainfall of each second monitoring area;

[0042] The adjustment module is configured to adjust the position of each second monitoring area based on the rainfall variation trend between adjacent second monitoring areas, and obtain an overlapping area of the second monitoring areas; when the rainfall increase path directions of the second monitoring areas are all directed to the overlapping area, the overlapping area is extracted as a rainfall center area;

[0043] The feature extraction module is configured to extract the rainfall-related features of the rainfall center area based on the position of the rainfall center area;

[0044] The output module is configured to receive the rainfall-related features of the rainfall center area based on the trained machine learning model, and output a short-impending heavy rain forecast result.

[0045] The above scheme has the following beneficial effects:

[0046] 1、The present application can reflect the rainfall characteristics of the entire target area by dividing the first monitoring area in the to-be-rainfall area, and the division of the second monitoring area can help to more accurately reflect the spatial variation of rainfall and provide a basis for subsequent determination of the rainfall center area.

[0047] Compared with the prior art, by comparing the rainfall data collected by the meteorological stations in each second monitoring area, the rainfall variation trend between adjacent second monitoring areas is finally obtained, the rainfall conditions of each position in the first monitoring area can be mastered, and the paths of rainfall increase or decrease of different sizes can be obtained, so that the system will not miss any rainfall when monitoring the rainfall, and the accuracy of the method is improved. At the same time, by more accurately obtaining the rainfall center, a more accurate short-impending heavy rain forecast result can be output.

[0048] 2、The application can find the position of the precipitation center area by adjusting the position of the second monitoring area between each adjacent second monitoring area in the first monitoring area according to the precipitation change trend, accurately determine the position of the precipitation center area in the first monitoring area by changing the size of the second monitoring area, and monitor some other precipitation points in the first monitoring area, so that the whole precipitation forecast is more comprehensive.

[0049] 3、The application can improve the accuracy of the forecast by determining the precipitation center area to predict heavy rain. At present, most of the traditional methods use radar data extrapolation to predict the rainfall area, and the position of the precipitation can be determined more accurately by determining the precipitation center area, which is more accurate than the traditional rainfall area prediction. After determining the precipitation center area, the numerical value of the precipitation center can be determined to achieve timing, quantitative and point prediction, improve the accuracy and pertinence of the precipitation forecast, provide heavy rain warning for the public and industry departments, avoid potential safety risks, and meet the urgent needs of the country for weather forecast refinement.

[0050] 4、The application can accurately identify the position, intensity and range of the precipitation center by extracting the features of the precipitation center area, more accurately predict the occurrence and evolution of precipitation, and establish a more refined prediction model by extracting and analyzing the features of the precipitation center, improve the spatio-temporal resolution and accuracy of the prediction. The precipitation center is generally a high-risk area of meteorological disasters such as heavy rain and flood, and extracting the features of these areas can help to discover disaster signs earlier, so as to predict the influence range and degree of the disaster and reduce the loss.

[0051] Additional aspects and advantages of the application will be described in part in the following description, some of which will become apparent from the following description, or will be learned by practice of the application. BRIEF DESCRIPTION OF DRAWINGS

[0052] Figure 1 The flowchart of the embodiment of the short-impending heavy rain forecast method of the application.

[0053] Figure 2 The structural block diagram of the embodiment of the short-impending heavy rain forecast system of the application. DETAILED DESCRIPTION

[0054] The technical solutions of the application will be described in detail below with reference to the drawings. Obviously, the described embodiments are part of the embodiments of the application, not all embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the application.

[0055] The specific embodiments will be described in detail below:

[0056] Embodiment 1:

[0057] As shown in the accompanying drawings: a short-approaching rainfall prediction method, comprising the following steps: Figure 1

[0058] S1: monitoring a rainfall area, and dividing the rainfall area into a first monitoring area after the rainfall area is monitored.

[0059] Wherein, the dividing of the rainfall area into the first monitoring area comprises: determining a monitoring purpose, collecting basic data, analyzing rainfall characteristics, and determining a monitoring area range. First, the specific purpose of monitoring the rainfall area needs to be clear, different purposes may require different monitoring strategies and data accuracy, then the basic geographic information of the area to be studied is collected to obtain historical meteorological data, including rainfall amount, rainfall frequency, rainfall type (such as rainfall, snowfall) and the like, according to the historical meteorological data, the rainfall characteristics of the target area are analyzed, including the spatial distribution of rainfall amount, the temporal variation, and the extreme rainfall event and the like, and finally according to the analysis result, the rainfall area is determined, and the rainfall area is divided into the first monitoring area.

[0060] S2: dividing the first monitoring area with the meteorological monitoring station as the center to obtain at least three second monitoring areas of regular shape, and all of which are located in the first monitoring area, and the area of the second monitoring area is smaller than that of the first monitoring area.

[0061] Wherein, the dividing of the first monitoring area with the meteorological monitoring station as the center to obtain the second monitoring area of regular shape (the regular shape can be circular, sector, rectangular or triangular, etc., and the embodiment selects circular) comprises:

[0062] S201: calculating the number of meteorological monitoring stations in the first monitoring area, the position distribution and the distance between adjacent meteorological monitoring stations by the meteorological monitoring station data collected by the computer to obtain a meteorological monitoring station distribution map of the first monitoring area;

[0063] S202: circularly dividing the first area based on the meteorological monitoring station distribution map of the first monitoring area, and the second detection area in the circular shape can cover the meteorological monitoring station distribution map of the first monitoring area;

[0064] S203: determining that each meteorological monitoring station is located at the center of the circle based on the second monitoring area in the circular shape in S202;

[0065] S204: determining the size of the second monitoring area by collecting the basic data of the second monitoring area through the computer;

[0066] S205: obtaining a second monitoring area distribution map of the first monitoring area based on S201-S204.

[0067] ​S3: Calculate the precipitation intensity distribution of different positions in the second monitoring area to which the meteorological monitoring station belongs and the total precipitation in the second precipitation area through the precipitation data of the same time period monitored by each meteorological monitoring station.

[0068] Specifically, it includes:

[0069] S301: Based on the range of each second monitoring area, the precipitation observation and radar detection are performed through the meteorological monitoring stations in the second monitoring area.

[0070] S302: Based on the precipitation observation and radar detection of the meteorological monitoring stations in each second monitoring area, the precipitation intensity and precipitation data of the second monitoring area are obtained.

[0071] S303: Analyze the precipitation intensity and precipitation data of each second detection area to obtain the precipitation intensity distribution in each second monitoring area and the total precipitation in the second precipitation area.

[0072] Among them, the precipitation intensity classification includes:

[0073] Light rain: the precipitation in 1 hour is less than or equal to 2.5mm;

[0074] Moderate rain: the precipitation in 1 hour is between 2.6mm and 8.0mm;

[0075] Heavy rain: the precipitation in 1 hour is between 8.1mm and 15.9mm;

[0076] Heavy rain: the precipitation in 1 hour reaches or exceeds 16mm.

[0077] S4: Determine the precipitation change trend in the second monitoring area and the precipitation change trend between adjacent second monitoring areas based on the precipitation intensity distribution and total precipitation of each second monitoring area.

[0078] Specifically, by analyzing the precipitation data in the second detection area, the precipitation change trend in the second monitoring area is obtained; by comparing the precipitation data of adjacent second monitoring areas, the size of the group of precipitation data is determined, and the change trend of the precipitation amount in the precipitation change area is obtained.

[0079] S5: Adjust the position of the second monitoring area based on the precipitation change trend between adjacent second monitoring areas, and obtain the overlapping area of the second monitoring area; when the precipitation increase path direction of each second monitoring area is towards the overlapping area of the second monitoring area, extract the overlapping area as the precipitation center area.

[0080] Specifically, it includes the following steps:

[0081] S501: Collect precipitation data of corresponding meteorological stations in each second monitoring area;

[0082] S502: Collect and aggregate the precipitation data of the weather stations in the second monitoring area within the first monitoring area;

[0083] S503: Compare the aggregated precipitation data of the weather stations, determine the precipitation trend and the path of precipitation increase or decrease;

[0084] S504: Determine whether the second monitoring area boundary is on the path of precipitation increase, and change the size and position of each second monitoring area to obtain the overlapping area of each second monitoring area;

[0085] S505: Determine whether the precipitation increase path direction of each second monitoring area is towards the overlapping area of the second monitoring area. If not, return to the precipitation data comparison unit, compare the precipitation data in the precipitation increase path direction, output a larger precipitation increase path, and adjust the second monitoring area again until the precipitation increase path direction of each second monitoring area is towards the overlapping area of the second monitoring area, and then extract the overlapping area as the precipitation center area.

[0086] S6: Based on the position of the precipitation center area, extract the precipitation-related features of the precipitation center area.

[0087] Specifically, based on S1 to S5, the specific position of the precipitation center area is determined, the precipitation data of the precipitation center area is collected, including historical data and real-time data; based on data analysis, the main factors affecting the precipitation characteristics of the precipitation center area are summarized, such as topography, climate, monsoon, etc.; the main precipitation characteristics of the precipitation center area (precipitation intensity, precipitation form, precipitation amount, environmental conditions (such as wind field, humidity, temperature, etc.), topographic information, etc.) are summarized.

[0088] S7: Based on the trained machine learning model, receive the precipitation-related features of the precipitation center area, and output the short-term rainfall forecast result.

[0089] Specifically, based on the main precipitation characteristics of the precipitation center area, the features are input into the machine learning model (Support Vector Machine (SVM), Random Forest, Deep Neural Network (DNN), etc.) that has been constructed and trained. Since in the existing machine learning model that has been constructed, the Random Forest model has the advantages of high prediction accuracy in ensemble learning, processing of nonlinear relationships and feature selection, etc., the precipitation-related features of the precipitation center area extracted in S6 are input into the Random Forest model, and then the Random Forest model can predict the rainfall map of the next moment of the precipitation center area, thereby obtaining the rainfall sequence of the future of the precipitation center area. Outputting the rainfall sequence can obtain the short-term rainfall forecast map of the precipitation center area.

[0090] The embodiment can improve the accuracy of the prediction by determining the precipitation center area to predict heavy rainfall. At present, the traditional method mostly uses radar data extrapolation to predict the rainfall area, and the position of the precipitation can be determined more accurately by determining the precipitation center area, which is more accurate than the traditional rainfall area prediction. After determining the precipitation center area, the numerical value of the precipitation center can be determined to achieve timing, quantification and point prediction, providing heavy rainfall warning for the public and industry departments, avoiding potential safety risks, and meeting the urgent needs of the country for weather prediction refinement. By determining the precipitation center area and combining the existing Random Forest model, the prediction accuracy can be better improved, and large-scale data sets and important features can be effectively processed and evaluated, thereby realizing the accuracy and pertinence of the short-term heavy rainfall prediction.

[0091] Embodiment 2:

[0092] As shown in the accompanying drawings: a short-term heavy rainfall prediction system based on the short-term heavy rainfall prediction method described in embodiment 1, comprising: a division module, a collection module, a judgment module, an adjustment module, a feature extraction module and an output module. Figure 2 A monitoring module is used to monitor the precipitation area, and when the precipitation area is monitored, the precipitation area is divided into a first monitoring area;

[0093] A division module is used to divide the first monitoring area with the meteorological monitoring station as the center to obtain at least three second monitoring areas of regular shape, which are all located in the first monitoring area, and the area of the second monitoring area is smaller than that of the first monitoring area;

[0094] A collection module is used to calculate the precipitation intensity distribution of different positions in the second monitoring area to which the meteorological monitoring station belongs and the total precipitation in the second precipitation area through the precipitation data of the same time period monitored by each meteorological monitoring station;

[0095] A judgment module is used to judge the precipitation intensity distribution and total precipitation of each second monitoring area to determine the precipitation change trend in the second monitoring area and the precipitation change trend between adjacent second monitoring areas;

[0096] An adjustment module is used to adjust the position of the second monitoring area based on the precipitation change trend between adjacent second monitoring areas, and obtain the overlapping area of the second monitoring area; when the precipitation increase path direction of each second monitoring area is towards the overlapping area of the second monitoring area, the overlapping area is extracted as the precipitation center area;

[0097] A feature extraction module is used to extract the precipitation-related features of the precipitation center area based on the position of the precipitation center area;

[0098] ​

[0099] An output module is configured to receive the precipitation-related features of the precipitation center area based on the trained machine learning model, and output the short-impending heavy rainfall prediction result.

[0100] Obviously, the above embodiments are only examples for clearly illustrating the present application and are not intended to limit the embodiments. Based on the above description, other different forms of changes or variations can be made by those skilled in the art. Here, all the embodiments are not required to be exhausted. The obvious changes or variations derived therefrom are still within the protection scope of the present application.

Claims

1. A method for short-term forecasting of heavy rainfall, characterized in that, Includes the following steps: S1: Monitor precipitation area. Once a precipitation area is detected, it will be designated as the first monitoring area. S2: Divide the first monitoring area with the meteorological monitoring station as the center to obtain at least three regular-shaped second monitoring areas, all of which are located within the first monitoring area, and the area of ​​the second monitoring area is smaller than the area of ​​the first monitoring area; S3: Calculate the distribution of precipitation intensity at different locations within the second monitoring area to which the meteorological monitoring station belongs and the total precipitation within the second monitoring area by using precipitation data monitored by each meteorological monitoring station for the same time period; S4: Based on the precipitation intensity distribution and total precipitation in each second monitoring area, determine the precipitation change trend within the second monitoring area and the precipitation change trend between adjacent second monitoring areas; S5: Adjust the location of the second monitoring area based on the precipitation change trend between adjacent second monitoring areas, and at the same time obtain the overlapping area of ​​the second monitoring area; when the precipitation increase path of each second monitoring area is towards the overlapping area of ​​the second monitoring area, extract the overlapping area as the precipitation center area; S6: Based on the location of the precipitation center region, extract precipitation-related features of the precipitation center region; S7: The training-based machine learning model receives precipitation-related features from the precipitation center area and outputs short-term heavy rainfall forecast results.

2. The method for short-term heavy rainfall forecasting according to claim 1, characterized in that, In S1, the precipitation area is divided into the first monitoring area, which includes: determining the monitoring purpose, collecting basic data, analyzing precipitation characteristics, and determining the scope of the monitoring area.

3. The method for short-term heavy rainfall forecasting according to claim 1, characterized in that, In S2, the first monitoring area is divided with the meteorological monitoring station as the center, resulting in at least three regularly shaped second monitoring areas, including: S201: Obtain the number and location of meteorological monitoring stations within the first monitoring area; S202: Determine the shape of the second monitoring area based on the range of the first monitoring area to ensure that the shape distribution can cover the first monitoring area; S203: Based on the number and location of meteorological monitoring stations within the first monitoring area, divide the second monitoring area into a regular-shaped area centered on the meteorological monitoring stations and determine the location of the second monitoring area; S204: Determine the size of the second monitoring area based on the shape and size of the second monitoring area.

4. The method for short-term heavy rainfall forecasting according to claim 1, characterized in that, S3: Calculate the precipitation intensity distribution at different locations within the second monitoring area to which each meteorological monitoring station belongs, and the total precipitation within that second monitoring area, using precipitation data from various meteorological monitoring stations over the same time period. This includes: S301: Radar detection is conducted through meteorological monitoring stations within each second monitoring area, based on the scope of each second monitoring area; S302: Based on radar detection by meteorological monitoring stations within each second monitoring area, obtain the precipitation intensity and precipitation data of that second monitoring area; S303: Analyze the precipitation intensity and precipitation data of each second monitoring area to obtain the precipitation intensity distribution and total precipitation in each second monitoring area.

5. The method for short-term forecasting of heavy rainfall according to claim 4, characterized in that, Analyze precipitation data from each of the second monitoring areas to determine the trend of precipitation changes, including: By analyzing precipitation data within the second monitoring area, the precipitation change trend within that area is obtained; by comparing precipitation data from adjacent second monitoring areas, the magnitude of this set of precipitation data is determined, thus obtaining the precipitation change trend in the precipitation change area.

6. The method for short-term forecasting of heavy rainfall according to claim 5, characterized in that, S5: Adjust the location of the second monitoring area based on the precipitation change trend between adjacent second monitoring areas, and at the same time obtain the overlapping area of ​​the second monitoring area; When the precipitation increase path of each second monitoring area is directed towards the overlapping area of ​​the second monitoring areas, this overlapping area is extracted as the precipitation center region, including: Based on the comparison of precipitation change trends within adjacent second monitoring areas, the paths of increasing or decreasing precipitation within these adjacent second monitoring areas are obtained. Through precipitation data analysis, the second monitoring areas within the precipitation change area will move towards areas with higher precipitation data along the path of increasing precipitation. Based on the movement of each second monitoring area, the overlapping areas of each second monitoring area are obtained. When the direction of increasing precipitation in each second monitoring area is towards the overlapping area of ​​the second monitoring areas, the overlapping area is extracted as the precipitation center area.

7. The method for short-term forecasting of heavy rainfall according to claim 6, characterized in that, In S5, the location of the second monitoring area is adjusted based on the precipitation change trend between adjacent second monitoring areas to find the location of the precipitation center area, including: S501: Collect precipitation data from the corresponding meteorological stations within each second monitoring area; S502: Collect and summarize precipitation data from meteorological stations in the second monitoring area within the first monitoring area; S503: Compare the precipitation data from the collected meteorological stations to determine the precipitation trend and the path of increase or decrease in precipitation. S504: Determine whether the edge of the second monitoring area is on the path of increasing precipitation, and change the size and position of each second monitoring area to obtain the overlapping area of ​​each second monitoring area; S505: Determine whether the precipitation increase path direction of each second monitoring area is towards the overlapping area of ​​the second monitoring area. If not, return to the precipitation data comparison unit, compare the precipitation data in each precipitation increase path direction, output a larger precipitation increase path, and then adjust the second monitoring area until the precipitation increase path direction of each second monitoring area is towards the overlapping area of ​​the second monitoring area. Then extract the overlapping area as the precipitation center area.

8. The method for short-term heavy rainfall forecasting according to claim 1, characterized in that, S6: Based on the location of the precipitation center area, extract the precipitation-related characteristics of the precipitation center area, including: clarifying the specific location of the precipitation center area based on S1 to S5; collecting precipitation data of the precipitation center area, including historical data and real-time data; analyzing the main factors affecting the precipitation characteristics of the precipitation center area based on the data; and summarizing the main precipitation characteristics of the precipitation center area.

9. The method for short-term forecasting of heavy rainfall according to claim 1, characterized in that, S7: The process of receiving precipitation-related features of the precipitation center area and outputting short-term heavy rainfall forecast results based on the trained machine learning model is as follows: Based on the main precipitation features of the precipitation center area, the features are input into the machine learning model that has been built and trained. Then the machine learning model can predict the rainfall amount of the precipitation center area at the next moment, thereby obtaining the future rainfall sequence of the precipitation center area. Outputting the rainfall sequence will yield the short-term rainfall forecast data of the precipitation center area.

10. A short-term heavy rainfall forecasting system, operating based on the short-term heavy rainfall forecasting method according to any one of claims 1-9, characterized in that, include: The monitoring module is used to monitor precipitation areas. Once a precipitation area is detected, it is designated as the first monitoring area. The division module is used to divide the first monitoring area with the meteorological monitoring station as the center to obtain at least three regular-shaped second monitoring areas, all of which are located within the first monitoring area, and the area of ​​the second monitoring area is smaller than the area of ​​the first monitoring area; The data acquisition module is used to calculate the distribution of precipitation intensity at different locations within the second monitoring area to which the meteorological monitoring station belongs, as well as the total precipitation within the second monitoring area, by using precipitation data monitored by each meteorological monitoring station for the same time period. The judgment module is used to judge the precipitation intensity distribution and total precipitation in each second monitoring area, as well as the precipitation change trend between adjacent second monitoring areas. The adjustment module is used to adjust the position of the second monitoring area based on the precipitation change trend between adjacent second monitoring areas, and at the same time obtain the overlapping area of ​​the second monitoring areas; When the direction of the precipitation increase path in each second monitoring area is towards the overlapping area of ​​the second monitoring areas, the overlapping area is extracted as the precipitation center area. The feature extraction module is used to extract precipitation-related features of the precipitation center region based on its location. The output module is used to receive precipitation-related features of the precipitation center area based on the trained machine learning model and output the short-term forecast results of heavy rainfall.

Images