A method and system for regional precipitation prediction based on radar echo

Through the combination of radar echo map and rain station data, combined with geographical feature partitioning and segmentation, the Self-attention ConvLSTM model is used to predict regional precipitation, which solves the problems of sparse distribution of rain stations and simple distance weight coefficient, and improves precipitation Quantity prediction accuracy and local heavy rainfall prediction ability.

CN118859213BActive Publication Date: 2024-12-03SICHUANG TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411321633.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-23
Publication Date
2024-12-03
Estimated Expiration
2044-09-23

AI Technical Summary

Technical Problem

The prior art relies on the distribution density and distance weight coefficient of rainfall stations in regional precipitation prediction, and ignores factors such as terrain and wind direction, resulting in insufficient accuracy.

Method used

By obtaining the actual precipitation measured by the radar echo map and the rainfall measurement station, performing grid division and analysis, combining the Z-R relationship to calculate the radar echo precipitation, and partitioning and segmenting according to geographical features, performing linear fitting, using Self-attention ConvLSTM The model predicts precipitation.

Benefits of technology

The accuracy of precipitation calculation has been improved, the ability to predict local heavy rainfall has been enhanced, the lack of reporting of station data has been supplemented, and the accuracy of water conservancy business analysis has been improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118859213B_ABST
    Figure CN118859213B_ABST
Patent Text Reader

Abstract

The present application relates to a method and system for regional precipitation prediction based on radar echo, comprising the following steps: obtaining radar echo images and measured precipitation at a rain gauge station in the same historical time period and the same area; parsing the radar echo images to obtain radar color spots at different times for each grid; calculating the rainfall intensity of the radar color spots through the Z-R relationship to obtain the radar echo precipitation of the grid where the rain gauge station is located; partitioning and segmenting the radar echo precipitation according to different geographical features and needs; performing linear fitting on the radar echo precipitation of different partitioned grids and the measured precipitation data of the rain gauge station in segments to output a fitting curve; obtaining a radar echo image to be tested, parsing the radar echo image to be tested, and calculating the radar echo precipitation through the fitting curve to obtain precipitation data of all grids in the region; using the calculated precipitation data as the input of a Self-attention ConvLSTM model to realize precipitation prediction in the region, and then predict the rain intensity.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of precipitation forecasting, and more specifically, to a method and system for regional precipitation prediction based on radar echoes. Background Art

[0002] Regional precipitation generally refers to the average precipitation over the entire region within a certain time frame. Precipitation data is the main factor in the simulation process of most water conservancy professional models. Mastering the regional precipitation situation and realizing the statistics and prediction analysis of regional precipitation can provide data support for regional water conservancy business. Therefore, it is of great significance to improve the accuracy of regional precipitation calculation.

[0003] As for regional precipitation, it is currently further calculated based on the existing high-precision rain monitoring station data through some specific methods. There are three main calculation methods. The first is the arithmetic mean method, which divides the precipitation of each station in the region by the number of rain gauges to get the average precipitation; the second is the Thiessen polygon method, which is a weighted average algorithm that takes into account the area of ​​the station area. It divides the calculation area into several polygonal units to ensure that there is a rain gauge near the center of each unit, so that the average precipitation in the region can be deduced based on the area occupied by each rain gauge; the third is the isohyetal method, which draws isohyetal lines, and then uses a planimeter or other methods to measure the area between adjacent isohyetal lines, and multiplies it by the average rain depth between the two isohyetal lines to get the precipitation on the area. Finally, the total precipitation on each part of the area is added up and divided by the full area to get the average precipitation in the region.

[0004] Prior art, such as the Chinese patent with patent number "CN108761576A", discloses a method and system for fusing X-band meteorological radar and rain gauge station data. The method includes: obtaining the rainfall data of the rain gauge station measured by the rain gauge station covered by the entire detection area of ​​the meteorological radar; obtaining the radar echo intensity data generated by the meteorological radar, gridding the radar echo intensity data, and obtaining the grid echo intensity data; inverting the grid echo intensity data into grid rainfall intensity data, and obtaining the original grid rainfall data in the entire rainfall process by the linear average accumulation method; searching for rain gauge stations within a certain range with the grid point as the center, and calculating the distance weight coefficient of the grid point according to the distance between the grid point and the searched rain gauge station; combining the distance weight coefficient, fusing the rainfall data of the rain gauge station measured by the rain gauge station at the grid point into the original grid rainfall data of the grid point, and obtaining the grid rainfall data after fusion of the grid point. The invention improves the accuracy of regional precipitation estimation by using the distance weight coefficient fusion method.

[0005] The problems existing in the above prior art are:

[0006] 1. Dependence on the distribution of rain gauges: The distribution density and location of rain gauges may be uneven and may be sparse in some areas. This may result in a limited number of reference rain gauges for some grid points when searching for rain gauges within a certain range, thus affecting the accuracy of the distance weight coefficient and the fusion effect.

[0007] 2. Limitations of distance weight coefficient: Calculating the weight coefficient based only on distance may be too simplistic. In actual situations, factors such as topography and wind direction may have an important impact on precipitation distribution, but they are not included in the calculation of distance weight coefficient. Summary of the invention

[0008] In order to solve the above problems existing in the prior art, the present invention proposes a method and system for regional precipitation prediction based on radar echo.

[0009] The technical solution of the present invention is as follows:

[0010] On the one hand, the present invention provides a method for regional precipitation prediction based on radar echo, comprising the following steps:

[0011] Step S1, obtaining radar echograms and measured precipitation at rain gauge stations in the same historical time period and the same area; dividing the area into grids, and analyzing the radar echograms to obtain radar color spots at different times for each grid;

[0012] Step S2, calculating the rainfall intensity of the radar color spot of each grid by using the ZR relationship to obtain the radar echo precipitation of the grid where the rain gauge station is located;

[0013] Step S3, dividing the radar echo precipitation into zones and segments according to different geographical features and needs;

[0014] Step S4, performing linear fitting on the radar echo precipitation of different partitioned grids and the measured precipitation data of the rain gauge station in the grids in sections; outputting fitting curves of the radar echo precipitation of different partitions and the measured precipitation;

[0015] Step S5, obtaining the radar echo map to be measured; analyzing the radar echo map to be measured and calculating the radar echo precipitation of all grids according to the corresponding fitting curve to obtain the precipitation data of all grids in the area;

[0016] Step S6, using the calculated precipitation data of all grids as the input of the Self-attention ConvLSTM model to realize the precipitation prediction in the area and then predict the rainfall intensity.

[0017] As a preferred implementation, the radar color spot of each grid is calculated for rainfall intensity through the ZR relationship, and the specific calculation formula is as follows:

[0018]

[0019] Where Z is the radar reflectivity factor, R is the radar echo precipitation, and a and b are empirical coefficients obtained through actual observation and statistical analysis.

[0020] As a preferred implementation, the radar echo precipitation is zoned and segmented according to different geographical features and needs; the zones are divided into mountainous areas, coastal areas, and plain areas according to different geographical features; and the segments are segmented according to different magnitudes of radar echo precipitation in different zones.

[0021] As a preferred implementation, the radar echo precipitation of different partition grids and the measured precipitation data of the measuring station in the grid are linearly fitted in sections; the specific fitting formula is as follows:

[0022]

[0023] Where S is the measured precipitation at the rain gauge station; R is the radar echo precipitation; x and y are fitting coefficients.

[0024] On the other hand, the present invention also provides a system for regional precipitation prediction based on radar echoes, comprising:

[0025] The data acquisition and processing module obtains the radar echo map and the measured precipitation of the rain gauge station in the same historical time period and the same area; divides the area into grids, and analyzes the radar echo map to obtain the radar color spots of each grid at different times;

[0026] The radar echo precipitation calculation module calculates the rainfall intensity of the radar color spot of each grid through the ZR relationship to obtain the radar echo precipitation of the grid where the rain gauge station is located;

[0027] The partition and segmentation module divides the radar echo precipitation into different zones and segments according to different geographical features and needs;

[0028] The linear fitting module performs linear fitting on the radar echo precipitation of different partition grids and the measured precipitation data of the rain gauge station in the grid in sections; outputs the fitting curves of the radar echo precipitation and the measured precipitation of different partitions;

[0029] The regional precipitation calculation module obtains the radar echo map to be tested, analyzes the radar echo map to be tested, and calculates the radar echo precipitation of all grids according to the corresponding fitting curve to obtain the precipitation data of all grids in the region;

[0030] The rain prediction module uses the calculated precipitation data of all grids as the input of the Self-attention ConvLSTM model to predict the precipitation in the area and then predict the rain intensity.

[0031] As a preferred implementation, the radar echo precipitation calculation module calculates the rainfall intensity of the radar color spot of each grid through the ZR relationship. The specific calculation formula is as follows:

[0032]

[0033] Where Z is the radar reflectivity factor, R is the radar echo precipitation, and a and b are empirical coefficients obtained through actual observation and statistical analysis.

[0034] As a preferred implementation, the partitioning and segmenting module partitions and segments the radar echo precipitation according to different geographical features and needs; the partitions are divided into mountainous areas, coastal areas, and plain areas according to different geographical features; and the segments are segmented according to different magnitudes of the radar echo precipitation in different partitions.

[0035] As a preferred implementation, the linear fitting module performs linear fitting on the radar echo precipitation of different partitioned grids and the measured precipitation data of the measuring stations in the grids in sections; the specific fitting formula is as follows:

[0036]

[0037] Where S is the measured precipitation at the rain gauge station; R is the radar echo precipitation; x and y are fitting coefficients.

[0038] On the other hand, the present invention further provides an electronic device having a computer program stored thereon, wherein when the computer program is executed by a processor, the method for regional precipitation prediction based on radar echo as described in any embodiment of the present invention is implemented.

[0039] On the other hand, the present invention also provides a computer-readable medium for storing one or more programs, which, when executed by the one or more processors, enables the one or more processors to implement a method for regional precipitation prediction based on radar echoes as described in any embodiment of the present invention.

[0040] The present invention has the following beneficial effects:

[0041] 1. Through the analysis results of radar echo images, in the case of missing or underreported data from measuring stations, precipitation monitoring data can be supplemented to improve the existing hydrological monitoring and forecasting system and provide a reference basis for water conservancy business analysis and judgment.

[0042] 2. Data curves with different characteristics can be fitted according to different partitions to obtain precipitation data that is more in line with regional characteristics, effectively improving the calculation accuracy of regional surface precipitation.

[0043] 3. The Self-attention ConvLSTM model uses precipitation data that has been partitioned and segmented, which can more accurately capture the spatiotemporal characteristics, improve the prediction ability of local heavy rainfall, and make the prediction results more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 This is a schematic diagram of the method flow of Example 1;

[0045] Figure 2 It is the point set diagram corresponding to the radar echo precipitation R and the measured precipitation S;

[0046] Figure 3 This is a schematic diagram of the overall structure of the Self-attention ConvLSTM model;

[0047] Figure 4 Schematic diagram of the SAM module structure in the Self-attention ConvLSTM model. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] It should be understood that the step numbers used in this document are only for convenience of description and are not intended to limit the order in which the steps are executed.

[0050] It should be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0051] The terms “include” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0052] The term "and / or" means and includes any and all possible combinations of one or more of the associated listed items.

[0053] Embodiment 1:

[0054] In order to make the purpose, technical scheme and advantages of the present invention more clear, the following will be combined with the specific embodiments of the present application and refer to the attached Figure 1 , clearly and completely describe the technical solution of the present invention.

[0055] This method processes the radar echo image and combines the ZR relationship to calculate the radar echo precipitation of each grid in the area. The measured precipitation data of the measuring station and the radar echo precipitation data are linearly fitted to obtain the measured precipitation and radar echo precipitation curve. Using this curve, the corresponding precipitation can be obtained through the radar echo precipitation of each grid, and the regional surface precipitation prediction is realized through the Self-attention ConvLSTM model, which expands the existing regional precipitation data source, improves the calculation accuracy of regional precipitation, enables the calculation of water conservancy professional models such as hydrology, hydrodynamics, and water resources, and improves the "monitoring, forecasting, early warning, rehearsal, and planning" capabilities of regional water conservancy services.

[0056] To solve the problems of the prior art, the present invention provides a method for regional precipitation prediction based on radar echoes, comprising the following steps:

[0057] Step S1, obtaining radar echograms and measured precipitation at rain gauge stations in the same historical time period and the same area; dividing the area into grids, and analyzing the radar echograms to obtain radar color spots at different times for each grid;

[0058] Step S101, dividing the area into 1km*1km grids;

[0059] Step S102, analyzing the radar echo map to obtain radar spots at different times for each grid.

[0060] Step S2, calculating the rainfall intensity of the radar color spot of each grid by using the ZR relationship to obtain the radar echo precipitation of the grid where the rain gauge station is located;

[0061] The rainfall intensity of each grid radar spot is calculated by ZR relationship. The specific calculation formula is as follows:

[0062]

[0063] Where Z is the radar reflectivity factor, R is the radar echo precipitation, and a and b are empirical coefficients obtained through actual observation and statistical analysis.

[0064] Step S3, dividing the radar echo precipitation into zones and segments according to different geographical features and needs;

[0065] Step S301, the regions are divided into mountainous areas, coastal areas, and plain areas according to different geographical features;

[0066] Step S302, segmentation is performed according to different levels of radar echo precipitation in different partitions. The specific segmentation is as follows:

[0067]

[0068] Step S4, linearly fitting the radar echo precipitation of different partition grids and the measured precipitation data of the rain gauge station in the grid in sections; outputting the fitting curves of the radar echo precipitation of different partitions and the measured precipitation; specifically comprising the following steps:

[0069] Step S401, obtaining a radar echo image within a period of time for analysis, and calculating the radar echo precipitation R of each grid in different partitions as the horizontal coordinate;

[0070] Step S402, obtaining the measured data of rainfall monitoring stations in different zones within the same time range, matching the corresponding grids according to the longitude and latitude of the rainfall monitoring stations, and taking the measured precipitation S of the corresponding grid as the ordinate;

[0071] Step S403, the point set corresponding to the radar echo precipitation R and the measured precipitation S at different times in all grids in different partitions can be obtained, such as Figure 2 As shown;

[0072] Step S404: According to the data scatter distribution, in order to obtain a higher precision fitting result, the radar echo precipitation R in different partitions is segmented according to different magnitudes, and the segmentation is as shown in step S302;

[0073] Step S405: Use a linear fitting method to fit the fitting coefficients x and y of different partitions and segments; the specific fitting formula is as follows:

[0074]

[0075] Where S is the measured precipitation at the rain gauge station; R is the radar echo precipitation; x and y are fitting coefficients;

[0076] Step S406: The fitting curves of different segments are merged and spliced ​​to obtain the fitting curve of each partition.

[0077] Step S5, obtaining the radar echo map to be measured; analyzing the radar echo map to be measured and calculating the radar echo precipitation of all grids according to the corresponding fitting curve to obtain the precipitation data of all grids in the area.

[0078] Step S6, using the calculated precipitation data of all grids as the input of the Self-attention ConvLSTM model to realize the precipitation prediction in the region and then predict the rainfall intensity;

[0079] Step S601, using the grid precipitation data of the region fitted 5 minutes at a time over the past 6 hours as input data for the Self-attention ConvLSTM model;

[0080] Step S602, calculating similarity based on the features of the current time step and the features of the memory unit to obtain aggregate features;

[0081] Step S603, using a gating mechanism to adaptively update the memory M so that the SAM can capture long-range dependencies in the spatiotemporal domain;

[0082] Step S604: the self-attention memory module (SAM) outputs the precipitation forecast and rain intensity measurement results in the area.

[0083] Embodiment 2:

[0084] This embodiment provides a system for regional precipitation prediction based on radar echoes, including:

[0085] The data acquisition and processing module obtains the radar echo map and the measured precipitation of the rain gauge station in the same historical time period and the same area; divides the area into grids, and analyzes the radar echo map to obtain the radar color spots of each grid at different times;

[0086] The radar echo precipitation calculation module calculates the rainfall intensity of the radar color spot of each grid through the ZR relationship to obtain the radar echo precipitation of the grid where the rain gauge station is located;

[0087] The partition and segmentation module divides the radar echo precipitation into different zones and segments according to different geographical features and needs;

[0088] The linear fitting module performs linear fitting on the radar echo precipitation of different partition grids and the measured precipitation data of the rain gauge station in the grid in sections; outputs the fitting curves of the radar echo precipitation and the measured precipitation of different partitions;

[0089] The regional precipitation calculation module obtains the radar echo map to be tested, analyzes the radar echo map to be tested, and calculates the radar echo precipitation of all grids according to the corresponding fitting curve to obtain the precipitation data of all grids in the region;

[0090] The rain prediction module uses the calculated precipitation data of all grids as the input of the Self-attention ConvLSTM model to predict the precipitation in the area and then predict the rain intensity.

[0091] Embodiment three:

[0092] This embodiment provides an electronic device having a computer program stored thereon. When the computer program is executed by a processor, a method for regional precipitation prediction based on radar echoes as described in any embodiment of the present invention is implemented.

[0093] Embodiment 4:

[0094] This embodiment provides a computer-readable medium for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement a method for regional precipitation prediction based on radar echoes as described in any embodiment of the present invention.

[0095] In the embodiments of the present application, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent the existence of A alone, the existence of A and B at the same time, and the existence of B alone. Among them, A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b and c can be represented by: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, c can be single or multiple.

[0096] Those of ordinary skill in the art will appreciate that the various units and algorithm steps described in the embodiments disclosed herein can be implemented in a combination of electronic hardware, computer software, and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0097] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0098] In several embodiments provided in the present application, if any function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory; hereinafter referred to as: ROM), random access memory (Random Access Memory; hereinafter referred to as: RAM), disk or optical disk, and other media that can store program codes.

[0099] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for regional precipitation prediction based on radar echo, characterized in that: The following steps are involved: Step S1, obtaining radar echograms and measured precipitation at rain gauge stations in the same historical time period and the same area; Divide the area into grids and analyze the radar echo map to obtain radar spots at different times for each grid; Step S2, calculating the rainfall intensity of the radar color spot of each grid by using the ZR relationship to obtain the radar echo precipitation of the grid where the rain gauge station is located; Step S3, dividing the radar echo precipitation into zones and segments according to different geographical features and needs; the zoning content includes dividing into mountainous areas, coastal areas, and plain areas according to different geographical features; the segmentation content includes segmenting according to different magnitudes of the radar echo precipitation in different zones; Step S4, performing linear fitting on the radar echo precipitation of different partitioned grids and the measured precipitation data of the rain gauge station in the grids in sections; outputting fitting curves of the radar echo precipitation of different partitions and the measured precipitation; Step S5, obtaining the radar echo map to be measured; analyzing the radar echo map to be measured and calculating the radar echo precipitation of all grids according to the corresponding fitting curve to obtain the precipitation data of all grids in the area; Step S6, using the calculated precipitation data of all grids as the input of the Self-attention ConvLSTM model to realize the precipitation prediction in the area and then predict the rainfall intensity.

2. The method for regional precipitation prediction based on radar echo according to claim 1, characterized in that: The radar color spot of each grid is used to calculate the rainfall intensity through the ZR relationship. The specific calculation formula is as follows: Where Z is the radar reflectivity factor, R is the radar echo precipitation, and a and b are empirical coefficients obtained through actual observation and statistical analysis.

3. The method for regional precipitation prediction based on radar echo according to claim 1, characterized in that: The radar echo precipitation of different partition grids and the measured precipitation data of the measuring station in the grid are linearly fitted in sections; the specific fitting formula is as follows: Where S is the measured precipitation at the rain gauge station; R is the radar echo precipitation; x and y are fitting coefficients.

4. A system for regional precipitation prediction based on radar echo, characterized in that: include: The data acquisition and processing module obtains the radar echo map and the measured precipitation of the rain gauge station in the same historical time period and the same area; Divide the area into grids and analyze the radar echo map to obtain radar spots at different times for each grid; The radar echo precipitation calculation module calculates the rainfall intensity of the radar color spot of each grid through the ZR relationship to obtain the radar echo precipitation of the grid where the rain gauge station is located; A partitioning and segmenting module is used to partition and segment the radar echo precipitation according to different geographical features and needs; the partitioning content includes dividing it into mountainous areas, coastal areas, and plain areas according to different geographical features; the segmenting content includes segmenting according to different magnitudes of radar echo precipitation in different partitions; The linear fitting module performs linear fitting on the radar echo precipitation of different partition grids and the measured precipitation data of the rain gauge station in the grid in sections; outputs the fitting curves of the radar echo precipitation and the measured precipitation of different partitions; The regional precipitation calculation module obtains the radar echo map to be tested, analyzes the radar echo map to be tested, and calculates the radar echo precipitation of all grids according to the corresponding fitting curve to obtain the precipitation data of all grids in the region; The rain prediction module uses the calculated precipitation data of all grids as the input of the Self-attention ConvLSTM model to predict the precipitation in the area and then predict the rain intensity.

5. The system for regional precipitation prediction based on radar echo according to claim 4, characterized in that: The radar echo precipitation calculation module calculates the rainfall intensity of the radar color spot of each grid through the ZR relationship. The specific calculation formula is as follows: Where Z is the radar reflectivity factor, R is the radar echo precipitation, and a and b are empirical coefficients obtained through actual observation and statistical analysis.

6. The system for regional precipitation prediction based on radar echo according to claim 4, characterized in that: The linear fitting module performs linear fitting on the radar echo precipitation of different partition grids and the measured precipitation data of the measuring station in the grid in sections; the specific fitting formula is as follows: Where S is the measured precipitation at the rain gauge station; R is the radar echo precipitation; x and y are fitting coefficients.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method for regional precipitation prediction based on radar echo as described in any one of claims 1 to 3 is implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for regional precipitation prediction based on radar echoes as claimed in any one of claims 1 to 3 is implemented.