Radar map-based quantitative analysis method and system for precipitation trend of administrative divisions at different levels

Abstract

The invention provides a radar map-based quantitative analysis method and system for rainfall trends of administrative divisions at all levels. The method comprises the following steps of: obtaining current and previous real-time radar maps at a plurality of times; according to the administrative division boundary file, respectively extracting radar echo grids of a preset-level administrative region from the real-time radar maps of the plurality of times so as to obtain rainfall data of each level of administrative division; calculating the real-time rainfall trend of each level of administrative division according to the rainfall data of each level of administrative division obtained by the real-time radar maps of the plurality of times; using the rainfall prediction neural network model to predict a forecast radar map of a plurality of times in the future, and performing data processing on the forecast radar map to obtain a future rainfall trend of each level of administrative division; and obtaining a final rainfall trend analysis result of each level of administrative division. According to the method, radar precipitation analysis based on the multi-level administrative division is realized, precipitation trend quantitative analysis of the multi-level administrative division is generated, and data support is provided for disaster prevention and reduction.

Description

TECHNICAL FIELD

[0001] The present application relates to the meteorological technical field, and particularly relates to a quantitative analysis method and system for precipitation trend of administrative divisions at all levels based on a radar map. BACKGROUND

[0002] Radar emission is a signal reflected back by electromagnetic waves in the atmosphere through precipitation particles, which is called radar echo (reflectivity). According to the signal strength reflected back by the precipitation particles received by the radar, a radar echo map can be made by using different colors. Generally speaking, the stronger the reflected signal is, the stronger the precipitation is, so the precipitation intensity and distribution of the weather system can be determined by the radar echo map.

[0003] By applying weather radar and analyzing radar echo maps, the intensity, location and movement of disaster weather such as typhoon, thunderstorm, rainstorm, squall line, hail and tornado can be understood in time, and timely warning can be made. It has good application effect in disaster prevention and disaster resistance, economic construction and national defense construction. However, the application and analysis of radar maps still have the following shortcomings: 1. The radar map analysis is not implemented to specific provinces, cities, districts and counties. At present, the analysis of radar maps is still in the stage of "seeing the map and speaking". Radar data uses a standardized grid as the basic analysis unit, which is not matched to the administrative division unit of province, city, district and county, so it is difficult to directly generate hierarchical precipitation statistical results that meet the needs of administrative management.

[0004] 2. The real-time and predicted radar maps do not form quantitative trend analysis results. At present, the time sequence change of the radar map forms a qualitative precipitation movement and precipitation amount increase and decrease trend result. The qualitative trend result cannot realize the quantitative analysis of the dynamic precipitation evolution characteristics of the administrative divisions at all levels, and lacks the ability to comprehensively analyze the average precipitation amount and duration of the administrative divisions at all levels. SUMMARY

[0005] In order to solve the above-mentioned problems of the prior art, the present application provides a quantitative analysis method and system for precipitation trend of administrative divisions at all levels based on a radar map, which realizes quantitative analysis of precipitation trend of multi-level administrative divisions.

[0006] In order to achieve the above-mentioned purpose, the technical scheme adopted by the present application is as follows: In a first aspect, the present application provides a quantitative analysis method for precipitation trend of administrative divisions at all levels based on a radar map, comprising: Step S1, obtaining real-time radar maps at current and previous multiple time points; Step S2, according to the administrative division boundary file, extracting the radar echo grid of the preset level administrative region from the multiple time live radar maps respectively, and obtaining the precipitation data of the administrative divisions of various levels according to the radar echo grid of the preset level administrative region; Step S3, calculating the real-time precipitation trend of the administrative divisions of various levels according to the precipitation data of the administrative divisions of various levels obtained from the multiple time live radar maps; Step S4, using the precipitation prediction neural network model to predict the future multiple time forecast radar maps, and processing the forecast radar maps according to the step S2 to obtain the future precipitation trend of the administrative divisions of various levels; Step S5, obtaining the final precipitation trend analysis result of the administrative divisions of various levels according to the real-time precipitation trend and the future precipitation trend.

[0007] The present application has the advantages that: taking the administrative division boundary file as the geographical information division basis, the data of the live radar map is extracted to obtain the precipitation data of the administrative divisions of various levels, and the radar precipitation analysis based on the administrative divisions of various levels is realized. Meanwhile, on the basis of the radar analysis result of the administrative divisions of various levels, the quantitative trend analysis of the administrative divisions of various levels is carried out. And the future precipitation trend is calculated in combination with the precipitation prediction neural network model. Finally, the precipitation trend quantitative analysis of the administrative divisions of various levels is generated by combining the above two, which provides data support for disaster prevention and reduction.

[0008] Optionally, in the step S2, obtaining the precipitation data of the administrative divisions of various levels according to the radar echo grid of the preset level administrative region comprises the following steps: When the radar echo grid is at the boundary of the two adjacent preset level administrative regions, the echo area ratio and the precipitation amount are distributed according to the grid area occupied by the preset level administrative regions on both sides of the boundary.

[0009] According to the above description, compared with the conventional scheme of directly dividing the grid into the largest area division, the grid data is distributed to each division according to the percentage of the area occupied by the division, the division boundary is described more accurately, and the more accurate administrative division precipitation data calculation result is generated.

[0010] Optionally, the precipitation prediction neural network model is a sa-convlstm neural network method, and the loss function TS is: ; In the formula, A represents the number of events that occur in prediction and observation, B represents the number of events that occur in prediction and do not occur in observation, C represents the number of events that do not occur in prediction and do not occur in observation, O represents the observed precipitation, and F represents the predicted precipitation.

[0011] According to the above description, the loss function is replaced by the TS score for describing the precipitation, and the TS score is improved in view of the defect that the observation and the forecast difference are not considered, so that the accurate description of the spatial and temporal distribution of the precipitation is realized in the network, and the more accurate precipitation prediction result is generated.

[0012] Optionally, the precipitation data comprises a precipitation area proportion, a precipitation position, an average precipitation intensity and a maximum precipitation intensity.

[0013] Optionally, in the step S2, the precipitation data of the administrative divisions at different levels is obtained according to the radar echo grid of the preset level administrative region, comprising the following steps: obtaining the precipitation data of the preset level administrative region according to the radar echo grid; attributing the precipitation data of the preset level administrative region according to the level relationship between the administrative divisions at different levels and the preset level administrative region, to obtain the precipitation data of the administrative divisions at different levels.

[0014] Optionally, in the step S2, the precipitation data of the preset level administrative region is obtained according to the radar echo grid, comprising the following steps: obtaining the precipitation position in the preset level administrative region according to the color group of the radar echo grid, taking the area proportion of each color group in the preset level administrative region as the precipitation area proportion, taking the maximum echo position as the precipitation center position of the preset level administrative region, calculating the precipitation intensity by using the Z-R relationship, and calculating the average precipitation intensity and the maximum precipitation intensity of each preset level administrative region.

[0015] Optionally, the real-time radar chart is 5-12 times, and the prediction radar chart is 12-30 times.

[0016] Optionally, the administrative division boundary file is an administrative division shp file.

[0017] Optionally, the preset level administrative region is a county-level administrative region.

[0018] In a second aspect, the application provides a radar chart-based quantitative analysis system for precipitation trends of administrative divisions at different levels, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the radar chart-based quantitative analysis method for precipitation trends of administrative divisions at different levels in the first aspect when executing the computer program.

[0019] The technical effects of the radar chart-based quantitative analysis system for precipitation trends of administrative divisions at different levels provided in the second aspect are referred to the related description of the radar chart-based quantitative analysis method for precipitation trends of administrative divisions at different levels provided in the first aspect. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 A flowchart of a radar-based quantitative analysis method of precipitation trends of administrative divisions at various levels according to an embodiment of the present application; Figure 2 A flowchart of a radar-based quantitative analysis method of precipitation trends of administrative divisions at various levels according to an embodiment of the present application; Figure 3 A flowchart of a radar-based quantitative analysis method of precipitation trends of administrative divisions at various levels according to an embodiment of the present application; Figure 4 A flowchart of a radar-based quantitative analysis method of precipitation trends of administrative divisions at various levels according to an embodiment of the present application;

[0021] Explanation of reference numerals: 1: A radar-based quantitative analysis system of precipitation trends of administrative divisions at various levels; 2: A processor; 3: A memory. DETAILED DESCRIPTION

[0022] In order to better understand the above technical solutions, exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided so that the present application can be understood more clearly and thoroughly, and the scope of the present application can be fully conveyed to those skilled in the art.

[0023] Embodiment one Please refer to Figures 1 to 3 A radar-based quantitative analysis method of precipitation trends of administrative divisions at various levels, comprising the steps of: Step S1, obtaining real-time radar maps at a current time and at a plurality of previous times.

[0024] As Figure 3 It can be seen that the real-time radar maps in this embodiment are 9 times in total, including real-time radar map (t-8), real-time radar map (t-7), …, real-time radar map (t). In other embodiments, the real-time radar maps can be 5-12 times.

[0025] Step S2, extracting radar echo grids of a preset level administrative region from the real-time radar maps at the plurality of times according to an administrative division boundary file, and obtaining precipitation data of administrative divisions at various levels according to the radar echo grids of the preset level administrative region.

[0026] In this embodiment, the administrative division boundary file is an administrative division shp file, and the preset level administrative region is a county-level administrative region, including a district, a county-level city, a county, etc.

[0027] In the embodiment, in step S2, obtaining the precipitation data of the administrative divisions of each level according to the radar echo grid of the administrative divisions of the preset level comprises the following steps. When the radar echo grid is at the boundary between two adjacent administrative divisions of the preset level, the echo area ratio and the precipitation amount are allocated according to the grid area occupied by the administrative divisions of the preset level on both sides of the boundary.

[0028] That is, in the embodiment, the grid data is allocated to the administrative divisions according to the percentage of the area occupied by different divisions, compared with the conventional scheme of directly dividing the grid to the division with the largest area, the embodiment describes the division boundary more accurately, and can generate more accurate calculation results of the administrative division precipitation data.

[0029] In the embodiment, in step S2, obtaining the precipitation data of the administrative divisions of each level according to the radar echo grid of the administrative divisions of the preset level comprises the following steps. Step S21, obtaining the precipitation data of the administrative divisions of the preset level according to the radar echo grid.

[0030] In the embodiment, the precipitation data includes the precipitation area ratio, the precipitation position, the average precipitation intensity and the maximum precipitation intensity. Then step S21 comprises the following steps: According to the color groups of the radar echo grid, the precipitation position in the administrative divisions of the preset level is obtained, the area ratio of each color group in the administrative divisions of the preset level is taken as the precipitation area ratio, the maximum echo position is taken as the precipitation center position of the administrative divisions of the preset level, the precipitation intensity is calculated by using the Z-R relationship, and the average precipitation intensity and the maximum precipitation intensity of each administrative division of the preset level are calculated.

[0031] Specifically, in the embodiment, 15-30, 30-45, 45-60 and >60 dBZ are taken as the green, yellow, red and purple echo color groups respectively.

[0032] Step S22, according to the level relationship between the administrative divisions of each level and the administrative divisions of the preset level, attributing the precipitation data of the administrative divisions of the preset level to obtain the precipitation data of the administrative divisions of each level.

[0033] Reference Figure 2 It can be seen that, in the embodiment, according to the prefectural administrative divisions and the provincial administrative divisions to which the county-level administrative divisions belong, the precipitation area ratio, the precipitation position, the average precipitation intensity and the maximum precipitation intensity of each prefectural administrative division and provincial administrative division are obtained respectively.

[0034] Step S3, calculating the real-time precipitation trend of the administrative divisions of each level according to the precipitation data of the administrative divisions of each level obtained from the live radar charts of multiple time points.

[0035] Specifically, the change trend values of the precipitation area proportion, the average precipitation intensity and the maximum precipitation intensity are obtained by comparing the precipitation area proportion, the average precipitation intensity and the maximum precipitation intensity of the current time t and the previous time t-1. According to the precipitation of each level of administrative division at the current time t, it is determined whether the precipitation is continuous, and the continuous duration of the current time t is obtained in combination with the previous continuous duration.

[0036] Step S4, using the precipitation prediction neural network model to predict the prediction radar maps of future multiple times, and processing the prediction radar maps according to step S2 to obtain the future precipitation trend of each level of administrative division.

[0037] In the embodiment, as shown in the figure, Figure 3 The prediction radar maps are 18 times, that is, the prediction radar maps (t+1), the prediction radar map (t+2), …, the radar map (t+18) of the future 18 times are predicted. In other embodiments, the prediction radar maps can be 12-30 times.

[0038] Specifically, the precipitation prediction neural network model of the embodiment is a sa-convlstm neural network method, and the traditional TS score is A / (A+B+C). The traditional TS score is not sensitive to different precipitation levels, and does not consider the score difference caused by different differences between predicted and observed precipitation. The loss function TS after improvement of the embodiment is: ; In the formula, A represents the number of events that occur in prediction and observation, B represents the number of events that occur in prediction and do not occur in observation, C represents the number of events that do not occur in prediction and do not occur in observation, O represents the observed precipitation, and F represents the predicted precipitation.

[0039] The above loss function accurately describes the spatial and temporal distribution of precipitation in the network, and generates more accurate precipitation prediction results.

[0040] Therefore, step S2 is repeated to traverse the 18 prediction radar maps to obtain the precipitation area proportion, the precipitation position, the average precipitation intensity and the maximum precipitation intensity of each level of administrative division at each prediction time. And the quantitative change of the precipitation trend of each prediction time is obtained.

[0041] Step S5, obtaining the final precipitation trend analysis result of each level of administrative division according to the real-time precipitation trend and the future precipitation trend.

[0042] Thus, according to the real-time precipitation trend and the future precipitation trend, the final precipitation trend analysis result of each administrative division is obtained. For example, the overall precipitation duration is obtained by combining the precipitation duration of the current time t in step S3 and the predicted precipitation duration after the current time t obtained by analyzing the 18 forecast times in step S4.

[0043] In summary, the present application has the following advantages: 1. The administrative division boundary file is used as the basis for geographical information division, and data is extracted from the real-time radar map to obtain precipitation data of each administrative division, thereby realizing radar precipitation analysis based on multi-level administrative division.

[0044] 2. On the basis of the multi-level administrative division radar analysis result, quantitative trend analysis of each administrative division is performed, and the future precipitation trend is calculated in combination with the precipitation prediction neural network model. Finally, the above two are combined to generate a quantitative analysis of the precipitation trend of the multi-level administrative division, thereby providing data support for disaster prevention and mitigation.

[0045] 3. The present application allocates grid data to each division according to the percentage of area occupied by different divisions for the grid of the division boundary, thereby more accurately describing the division boundary and generating more accurate administrative division precipitation data calculation results.

[0046] 4. The present application replaces the loss function with an improved TS score that describes precipitation, thereby accurately depicting the spatial and temporal distribution of precipitation in the network and generating more accurate precipitation prediction results.

[0047] Embodiment Two Please refer to Figure 4 The multi-level administrative division precipitation trend quantitative analysis system 1 based on the radar map comprises a memory 3, a processor 2, and a computer program stored on the memory 3 and executable on the processor 2, wherein the processor 2 implements the steps in Embodiment One or Two when executing the computer program.

[0048] The system / device described in the above embodiments of the present application is the system / device used to implement the method of the above embodiments of the present application, and therefore, based on the method described in the above embodiments of the present application, those skilled in the art can understand the specific structure and modifications of the system / device, and therefore, it is not repeated here. Any system / device used in the method of the above embodiments of the present application belongs to the scope of protection of the present application.

[0049] Those skilled in the art will appreciate that embodiments of the present application can be readily used as a method, apparatus or computer program product. Accordingly, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application can take the form of a computer program product on one or more computer readable storage media (including, but not limited to, disk memory, CD-ROMs, optical storage devices, etc.) embodying computer readable program code.

[0050] The present application is described in reference to the flowchart illustrations and / or block diagrams of the methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a computer, or other programmable data processing apparatus, to produce a machine such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart illustrations and / or block diagrams.

[0051] It should be noted that the description of the application is not limited to the embodiments described above. It will be apparent to those skilled in the art that various modifications and variations can be made to the specific embodiments without departing from the scope or spirit of the application. One of the skills in the art will be apparent that the embodiments of the present application can be provided as a method, apparatus or computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product embodied on one or more computer usable storage media (including, but not limited to, disk memory, CD-ROM, optical storage, etc.) containing computer usable program code.

[0052] In addition, it should be noted that in the description of the specification, the description of the terms "one embodiment", "some embodiments", "embodiment", "example", "specific example" or "some examples" and the like means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In the specification, the illustrative description of the above terms is not necessarily directed to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, those skilled in the art can combine and combine the different embodiments or examples described in the specification and the features of the different embodiments or examples without contradiction.

[0053] Although the preferred embodiments of the present application have been described, those skilled in the art can make further changes and modifications to the embodiments after learning the basic creative concept. Therefore, the claims should be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0054] It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the spirit or scope of the application. Thus, it is intended that the present application cover modifications and variations of this application provided they come within the scope of the appended claims and their equivalents.

Claims

1. A method for quantitative analysis of precipitation trends in administrative divisions at all levels based on radar charts, characterized in that, include: Step S1: Obtain the real-time radar images for the current and previous time periods; Step S2: Based on the administrative division boundary file, extract the radar echo grid of the preset level administrative division from the real-time radar images of multiple time periods, and obtain the precipitation data of each level of administrative division based on the radar echo grid of the preset level administrative division. Step S3: Calculate the real-time precipitation trend of each administrative division based on the precipitation data of each administrative division obtained from the real-time radar images at multiple times. Step S4: Use a precipitation prediction neural network model to predict the forecast radar images for multiple future time periods, and process the forecast radar images according to step S2 to obtain the future precipitation trends of administrative divisions at all levels. Step S5: Based on the real-time precipitation trend and the future precipitation trend, obtain the final precipitation trend analysis results for each level of administrative division.

2. The method for quantitative analysis of precipitation trends in administrative divisions at all levels based on radar charts according to claim 1, characterized in that, In step S2, obtaining precipitation data for each level of administrative division based on the radar echo grid of the preset administrative division includes the following steps: When the radar echo grid is located at the boundary of two adjacent preset level administrative regions, the echo area ratio and precipitation are allocated according to the grid area occupied by the preset level administrative regions on both sides of the boundary.

3. The method for quantitative analysis of precipitation trends in administrative divisions at all levels based on radar charts according to claim 1, characterized in that, The precipitation prediction neural network model is the sa-convlstm neural network method, and its loss function TS is: ； In the formula, A represents the number of events that were predicted to occur and were observed to occur, B represents the number of events that were predicted to occur but were not observed to occur, C represents the number of events that were neither predicted to occur nor were observed to occur, O represents the observed precipitation, and F represents the predicted precipitation.

4. The method for quantitative analysis of precipitation trends in administrative divisions at all levels based on radar charts according to claim 1, characterized in that, The precipitation data includes the percentage of precipitation area, precipitation location, average precipitation intensity, and maximum precipitation intensity.

5. The method for quantitative analysis of precipitation trends in administrative divisions at all levels based on radar charts according to claim 4, characterized in that, In step S2, obtaining precipitation data for each level of administrative division based on the radar echo grid of the preset administrative division includes the following steps: The precipitation data of the preset administrative region is obtained based on the radar echo grid. Based on the hierarchical relationship between administrative divisions at all levels and the preset level of administrative divisions, precipitation data of the preset level of administrative divisions are assigned to obtain precipitation data for each level of administrative division.

6. The method for quantitative analysis of precipitation trends in administrative divisions at all levels based on radar charts according to claim 5, characterized in that, In step S2, obtaining the precipitation data of the preset level administrative region based on the radar echo grid includes the following steps: The precipitation location within the preset administrative level region is obtained based on the color group of the radar echo grid. The area ratio of each color group within the preset administrative level region is taken as the precipitation area ratio. The location of the maximum echo is taken as the precipitation center location of the preset administrative level region. The precipitation intensity is calculated using the ZR relationship. The average precipitation intensity and maximum precipitation intensity of each preset administrative level region are also calculated.

7. The method for quantitative analysis of precipitation trends in administrative divisions at all levels based on radar charts according to any one of claims 1 to 6, characterized in that, The real-time radar images are taken over 5 to 12 time periods, and the forecast radar images are taken over 12 to 30 time periods.

8. The method for quantitative analysis of precipitation trends in administrative divisions at all levels based on radar charts according to any one of claims 1 to 6, characterized in that, The administrative division boundary file is an administrative division shp file.

9. The method for quantitative analysis of precipitation trends in administrative divisions at all levels based on radar charts according to any one of claims 1 to 6, characterized in that, The preset administrative level is the county-level administrative region.

10. A quantitative analysis system for precipitation trends in administrative divisions at all levels based on radar charts, 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 computer program, it implements the quantitative analysis method for precipitation trends of administrative divisions at all levels based on radar charts as described in any one of claims 1 to 9.

Images