KR20100106745A - Radar observational density analysis system and analysis method of the same - Google Patents

Abstract

PURPOSE: A radar observational density analysis system and an analysis method of the same are provided to efficiently radar measurement data by accurately measuring quantitative precipitation depending on geographical factors. CONSTITUTION: A comprehensive information processor(100) manages the basic information and analysis information of a measurement point of a radar site, and a vector data generator(200) generates vector data according to the processed data of the comprehensive information processor. A three-dimensional image data generator(300) implements the generated vector data in three-dimensional image. An image analysis unit(400) analyzes the weather information through the analysis of the information implemented in three-dimension. A data display unit(500) displays the analyzed weather information.

Description

Radar observational density analysis system and analysis method of the same}

The present invention relates to a technique for analyzing spatial observation density of radar observation data and calculating QPE reliability. In particular, the radar data observation density analysis display system performs a comparative analysis of the correlation coefficient (R ² ) and the RMSE, which are the QPE reliability and radar precipitation accuracy when the user arbitrarily analyzes a specific region, and observes each observation altitude. By measuring the density D and calculating the observation altitude with high QPE reliability, the present invention relates to a technique for implementing an efficient system.

Rainfall estimation based on weather radar observations has been studied in depth. The application of various hydrologic applications such as heavy rain, snow, hail, heavy snow and floods, river flows, and fresh water according to weather changes, from the display of terrain with accurate and precise measurements of rainfall and hourly rainfall based on radar data. Provide essential predictive information about the study. The Korea Meteorological Administration has applied the cloud layer ZR relation of Z = 200R1.6 for all precipitation cases since 2005, but this relation was not appropriate due to convective rainfall in summer in Korea. A system has been developed to estimate rainfall by calibrating radar data. The system is called "Radar-AWS Rainrate (RAR) calculation system".

This RAR system uses the WPMM (Window Probability Matching Method; Rosenfeld et al., 1994) theory on the reflectivity observed by the Korea Meteorological Office and the rainfall measured by the rain gauge to determine the ZR relation to estimate the rainfall by radar. Rainfall over the Korean Peninsula is represented.

In particular, the precipitation in East Asia is closely related to the precipitation on the Korean Peninsula. In particular, in order to improve the monitoring and forecasting ability of heavy rain related to bad weather, quantitative precipitation estimation (QPE) in East Asia and the characteristics of the precipitation system are required. However, the wide distribution of oceans in this region is also one of the factors that makes it difficult to pinpoint accurate precipitation.

Precipitation in East Asia by using observation data from AWS ground observation-Korea-China radar- geostationary satellite-polar orbit microwave satellite data with different time / space scales to overcome the difficulties of observation in these regions and diversify the monitoring elements It is essential to develop the base technology on the limitations of data for calculating the chapter.

Therefore, by utilizing the comprehensive radar data of Korea-China, we monitor the bad weather from the Korean Peninsula to East Asia, and expand the capability of forecasting to concentrate on weather changes in South Korea. And the development of a system capable of performing the analysis accordingly.

The present invention has been made to solve the above-described problems, the object of the present invention is to solve the problem of difficult to accurately grasp the precipitation according to the geographical environmental factors, to overcome the difficulties caused by the environmental factors of the observation and monitoring of the precipitation plant It is possible to diversify the elements, thereby providing a system that can more efficiently calculate East Asian precipitation.

In particular, by utilizing the Korea-China radar data comprehensively, we can monitor the bad weather from the Korean Peninsula to East Asia and implement a system that can intensively predict weather changes to efficiently analyze and express the observation density of radar observation data for specific areas. The purpose is to provide a system.

The present invention provides a means for solving the above problems, Comprehensive information processing unit for managing the basic information and analysis information of the observation point of the radar site; A vector material generation unit generating vector data based on the data processed by the comprehensive information processing unit; A three-dimensional image data generation unit for implementing the three-dimensional image of the vector material generated by the vector material generation unit; An image analyzer for analyzing weather information by analyzing the information implemented as the 3D image; A data display unit for displaying the analyzed weather information; to provide a radar observation density analysis system comprising a.

In particular, the comprehensive information processing unit in the present invention, the radar data analysis module for analyzing radar data and display radar data; A radar site information processing module for processing information of the radar site; Characterized in that comprises a, and further comprises a data management module for storing the information of the radar site and the data management module for the management of the information of the radar site.

The information processed and managed by the comprehensive information processor may include radar site information, correlation information, prediction error (RMSE), and radar data.

The radar observation density analysis method using the radar observation density analysis system described above may be performed in the following steps.

Displaying a map of a specific area and image information of precipitation observation of a radar; Displaying radar site-specific information of the specific region; Selecting an observation angle of the radar to express an observation radius grid point; 4 steps of expressing the precipitation distribution of the observation radius and the image information on the radar observation grid; 5 steps of analyzing the image information to calculate the analysis information; It may be implemented to include.

In the above-mentioned steps, the second step may adjust the information by inputting the position, height, radius, and observation angle of the radar site, and further, in the third step, in displaying the observation radius lattice point, the radar site may be adjusted. The angle of the star is specified, and the altitude of the observation point and the radar altitude are calculated.

In addition, the analysis information of the five steps in the present invention is implemented by calculating the comparative analysis of the quantitative precipitation estimation (QPE) reliability, correlation coefficient (R2), prediction error (RMSE) for a specific region Can be.

According to the present invention, it is possible to solve the problem that it is difficult to accurately grasp the precipitation according to the geographical environmental factors, to overcome the difficulties caused by the environmental factors of the observation and to diversify the monitoring elements of the precipitation plant, thereby making it possible to view the East Asian precipitation plant. There is an effect of providing a system that can be efficiently calculated.

In particular, by utilizing the Korea-China radar data comprehensively, we can monitor the bad weather from the Korean Peninsula to East Asia and implement a system that can intensively predict weather changes to efficiently analyze and express the observation density of radar observation data for specific areas. To provide a system.

Hereinafter, with reference to the accompanying drawings will be described a specific configuration according to the present invention and the embodiment applied thereto.

1 is a block diagram of a radar observation density analysis system according to the present invention.

The radar observation density analysis system according to the present invention generates a vector data based on the data processed by the integrated information processing unit 100 and the integrated information processing unit for managing the basic information and analysis information of the observation point of the radar site The unit 200, the three-dimensional image data generation unit 300 for implementing the vector data generated by the vector data generating unit 300, the image analysis to analyze the weather information by analyzing the information implemented in the three-dimensional image The unit 400 and a data display unit 500 for displaying the analyzed weather information is configured.

In particular, the comprehensive information processing unit 100 includes a radar data analysis module 110 for data processing and a radar site information processing module 120 for reading and processing various information on the radar site, wherein the radar site information Or a data management module 130 for managing the correlation, prediction error (RMSE), radar data, etc. at high speed, and a data storage module for storing a part of the data for the high-speed processing described above and managing the same. It is preferable that it further comprises 140).

The Radar Data Parsing module 110 is a radar data analysis module for data processing, and includes radar for data processing and display such as correlation information and root mean square error (RMSE). Process the data. Here, the average error is calculated by taking the square root of the difference between the predicted error (RMSE) and the observed value, and has a value ranging from '0' to infinity and '0' in a perfect case.

In addition, the radar site information processing module (Site Information module) 120 means a module for reading and processing a variety of information on the radar site (Radar Site) of a specific area that requires weather analysis, the user is to each site (Site) It is possible to modify the information on the site and to simulate the change of the site value.

The storage management module 130 is a module that manages the radar site information, Correlation, RMSE, and radar data to be processed at high speed by the system. The data can be managed by storing and loading data.

The data storage module 140 stores some of data in a main memory for high-speed processing of the data, and efficiently manages the data.

The vector graphic generation unit 200 is a module for generating and managing preprocessing vector data for processing the data processed in this system based on most vector data. .

In addition, the 3D image data generation unit 300 (3D Lib. Module) performs a function to implement the 3D image data by processing the 3D vector data generated by the vector data generation unit. Specifically, since the data generated by the vector data generation unit is a simple vector data, these vector data and raster data can be converted into actual data on a 3D volume according to the characteristics of meteorology, especially radar data. It is to perform the function of creating virtually.

The image analysis unit 400 is a module that actually implements a corresponding analysis in the information implemented on the 3D virtual volume by the 3D image data generating unit 300, so that accurate box counting is possible. The analysis is performed by applying the analysis methods such as the cross collision detection for the position and the box position and the collision detection with the radar data display area. In particular, it is desirable to implement a module for calculating Detected Point data.

The data display unit 500 is a module for providing a user with the final data processed and analyzed in the above-described configuration, and when implemented on a user terminal, defines a sub domain through a mouse. This information can then be provided to the analysis module. The displayed data can be enlarged and reduced by Zoom In / Out, and the area movement can be done by Panning. In particular, the site where the analysis is performed on the system (Site) can be arbitrarily selected by the user and the data can be displayed, and the above-described elevation angles can be arbitrarily selected by the user as described above.

Figure 2 shows the structure of the directory of the radar observation density analysis system in performing the operation of the analysis system according to the present invention, the contents of this structure will be described through Table 1 below.

{Table 1}

When the radar observation density analysis system according to the present invention is driven, the data is processed in the data, image, map, and setup data in the above-mentioned directory, the data is displayed in the 3D Virtual Voulme, and processed to express the data on the screen. Done.

In addition, radar-related data is processed to be displayed on 3D Virtual Volume after analysis by Radar Data Processing. This includes data such as correlation and RMSE.

The radar site information data processes the basic information of the site displayed on the 3D Virtul Volume. This enables simulation through a series of tasks, such as adding additional sites and adjusting the Theta value .

When the radar observation density analysis system according to the present invention is driven, the system is driven and the read data is processed and analyzed by the comprehensive information processing unit and then transferred from the data management module 130 to the data storage module 140 when necessary or when necessary. Disk cash is processed to ensure that the data is processed immediately when there is a need inside the system. In particular, some data needs to be processed quickly, so the data is processed in main memory or graphics memory.

In particular, vector data is an essential form for 3D graphics processing, and all data except raster data undergo Vector processing. This ensures that the data displayed on the screen always provides the best resolution, even when zooming in and out.

In addition, since the vector data is only simple data, it must be reprocessed into a data type suitable for the characteristics of meteorology, especially radar, etc., which is processed through the 3D image data generation unit 300 Data is displayed.

The vector and raster data generated by the 3D image generating unit are finally implemented on the 3D virtual volume. This provides a method that can be analyzed by the analysis module (image segmenter) in the same way as the reality in the information implemented on the 3D Virtual Volume. In addition, since data should be represented on a map, GIS map information is also implemented.

The image analyzer 400 analyzes the data in the 3D Virtual Volume in the same way as in the reality. To do this, the information must be processed in the information implemented on the GIS information. In particular, in order to apply the same method as Box Counting, real-time zeroing on the grid, box crossing collision detection in the selected area, and radar data display area and collision detection with the real area must be performed together. In addition, the data can be analyzed like the real world in response to changes in the radar altitude, and the situation according to the changes in theta values is also considered.

The final analyzed area is displayed on the user's screen. In addition, after calculating the sub domain designated by the user to match the domain of the real world, it is used to feed the user's interaction information back to the analysis module so that the data can be analyzed in the analysis module. Back)

Hereinafter, a practical application of the driving process of the system according to the present invention described above will be described with reference to the drawings.

The program execution screen in the basic user terminal (PC) will be described for each function as follows.

3A illustrates a screen when a system is driven to execute a program. We will secure the target area by catching the area with the Chinese and Japanese areas centered on the Korean Peninsula. (In particular, the preferred embodiment of the present invention implements a 1km * 1Km Grid including the center of the Korean Peninsula, China and Japan, and uses the standard map projection of KMA (Korea Meteorological Administration). As a map projection method for the data grid, Lambert Conic Conformal map Projection assumed the earth radius of 6371.00877 km. The measured distance correspondence latitude was set to 30, 60 ° N north latitude as an example.)

FIG. 3B illustrates that an arbitrary area is selected, and a setting box (BOX) for setting an arbitrary area is designated, which can be set with a mouse, and can be moved or adjusted in size. In addition, the user may enlarge or reduce the screen (FIG. 3C), or move the screen to an arbitrary area desired by the user (FIG. 3D). This operation can be controlled via the mouse.

3E shows a line map, and it is possible to designate whether the map is displayed (on / off) in the control option box window. In this system, this map can be selectively expressed as 30m precision line map and DEM map. Figure 3f shows the appearance to the DEM map. In addition, color expression according to altitude can be additionally set.

As shown in FIG. 3G, the latitude and longitude can be displayed in the present system, and in this case, the density can be adjusted according to the zoom situation. In particular, as shown in Figure 3h, when the grid is expressed, the grid for the observation point check (Detect Point Check) will be displayed, and the density can also be adjusted according to the zoom situation.

Shown in Fig. 3i is a selection window for selecting the site of the radar. That is, it is possible for the user to select a site to be displayed, and as shown in FIG. 3J, it is also possible to add an arbitrary site if necessary. This allows you to select an arbitrary point and calculate the data, if necessary, and to set the latitude and longitude. The data can be in the form of text, and preliminary checks can be made as the site is added or moved.

3k illustrates a process of selecting and expressing an elevation angle in the present system. That is, the elevation angle of the corresponding site can be easily selected, and the screen for the selected elevation angle is shown in FIG. 3L.

3m illustrates a screen for displaying an overlapping display of observation data of a radar. Select observations to display in a nested manner, and the observations will only display the contents kept in a specific directory. Of course, observation data may be KMA user fees, or may include compressed binary data, short integers, Lambert Conformal Conic, and 3 * 3 km resolution.

In particular, this system makes it possible to express the variable altitude of CAPPI. Referring to FIG. 3N, the CAPPI altitude setting shows that the base is set to 1 km, the top is 3 km, and the increment is 0.1 km, which can be adjusted by controlling the mouse and can be displayed in real time. As the basic display of radar data, the antenna altitude angle is divided into 18 from 0 ° to 45 ° to scan the data, and the basic value that displays the data once scanned is called PPI, especially CAPPI (Constant Altitude) Plan Position Indicator) is calculated by calculating each PPI as a plane value at a constant altitude.)

3o shows the count detected points. The user can specify the information of the subdomain, for example, observation density, average RMSE, average correlation coefficient, information of the selected area (latitude / longitude), or site information (such as site name-elevation-detected no. ) Can be specified. FIG. 3P shows the result of the above-described adjustment of the subdomain in FIG. 3O and the value in the adjusted domain.

3q detects and calculates a detected line and a box collision, and a count point is represented by a small rectangular box (blue) on the right side of the drawing.

Figure 3r shows the display of the name of the branch of the AWS (automatic weather observation system; automatic weather system), which also represents the main AWS point name, can be provided with information of the observation point, which can be operated by the mouse control Of course.

Figure 3s shows a process of performing management for each analysis case (case), it is possible to separate and store the work environment for each analysis case, the stored work environment can be recalled later or classified storage. The final working environment is automatically saved and can be used.

Figure 3t shows an example of applying the system to the information analysis of the unobserved area on the West Sea. The right figure "1" calculates the correlation between the precipitation of the observed area and the QPE calculated by the radar, and the left figure "2" shows the QPE reliability by the radar information where the irrigation observation is impossible. It is shown that is calculated. 3u illustrates a three-dimensional virtual image to which the present system is applied.

The process of analyzing weather information of a specific region will be briefly described with reference to application examples of each function of the present system.

(Example of application of meteorological information analysis in specific regions of East Asia)

First, the system selects an East Asian region from the above-described screen and displays a map of the region and radar precipitation observation image. Of course, in this process, the screen can be moved by dragging and dropping the mouse and zooming in and out of the image using the mouse wheel. It is possible.

Next, information on each radar site is displayed. The radar site information can be changed to a file such as site location, height, radius, and observation angle. If the program is restarted after the change, the content is applied to the program.

 Then, the radar observation angle is selected to express the observation radius grid point. Specifically, when the site is selected, the angle for each site is expressed, and when the desired angle is selected, the radar observation radius is expressed. (The radius is calculated by considering the altitude of the observation point and the radar altitude. In this system, the radius is expressed as 360 points in 1 degree intervals. The radius of observation represented by 360 points is 1km when the Grid option is checked. A grid with a radius of × 1km is represented, and it is possible to check not only whether the grid contains points but also where the area of the 1km × 1km area is zoomed in using the mouse wheel.)

  According to the above process, the information on the specific region can express various image information about the East Asian precipitation distribution and the radar observation grid by the system according to the present invention described above. When selecting East Asian radar observation data, it is displayed superimposed on the map information. When selecting Grid, AWS station, DEM, Lon / Lat, Line Map option, each information is displayed.

  Finally, radar observation density and QPE reliability information are displayed. In addition, when selecting an analysis area, information on 'observation density', 'average RMSE', 'average CORR', 'area information', and 'number of observation points per site angle' of the area are displayed. Of course, the analysis area can be changed by the user arbitrarily.

Through the above process, the spatial observation density of radar observation data can be analyzed and QPE reliability can be calculated. Specifically, the radar data observation density analysis expression system can be used to analyze the QPE reliability and radar precipitation accuracy. A comparative analysis of the correlation coefficient (R ² ) and RMSE is performed, and the observation altitude D of each observation altitude is measured to calculate the observation altitude with high QPE reliability, thereby enabling more efficient weather analysis.

In addition, the application examples described above are applied by calculating and expressing the observed density and the estimated radar precipitation performance (R2) for an arbitrary area or an empty space (the area without an ocean or rain gauge) by using the quantitative precipitation estimation reliability and the observation density relationship. It is also possible.

In the foregoing detailed description of the present invention, specific examples have been described. However, various modifications are possible within the scope of the present invention. The technical idea of the present invention should not be limited to the embodiments of the present invention but should be determined by the equivalents of the claims and the claims.

1A and 2B show the configuration of a radar observation density analysis system according to the present invention.

2 illustrates the structure of a directory of the analysis system according to the present invention.

3A to 3U are functional state diagrams showing a driving example of the analysis system according to the present invention for each function.

Claims (9)

Comprehensive information processing unit for managing the basic information and analysis information of the observation point of the radar site; A vector material generation unit generating vector data based on the data processed by the comprehensive information processing unit; A three-dimensional image data generation unit for implementing the three-dimensional image of the vector material generated by the vector material generation unit; An image analyzer for analyzing weather information by analyzing the information implemented as the 3D image; A data display unit displaying the analyzed weather information; Radar observation density analysis system, characterized in that comprises a. The method according to claim 1, The comprehensive information processing unit, A radar data analysis module for analyzing radar data and analyzing radar data for display; A radar site information processing module for processing information of the radar site; Radar observation density analysis system, characterized in that comprises a. The method according to claim 1 or 2, The comprehensive information processing unit, A data management module for managing information of the radar site; A data storage module for storing information of the radar site; Radar observation density analysis system, characterized in that further comprises. The method according to claim 3, Information processed and managed in the comprehensive information processing unit, Radar observation density analysis system comprising radar site information, correlation information (correlarion), prediction error (RMSE), radar data. Displaying a map of a specific area and image information of precipitation observation of a radar; Displaying radar site-specific information of the specific region; Selecting an observation angle of the radar to express an observation radius grid point; 4 steps of expressing the precipitation distribution of the observation radius and the image information on the radar observation grid; 5 steps of analyzing the image information to calculate the analysis information; Observation density analysis method of the radar comprising a. The method according to claim 5, The second step, Radar observation density analysis method, characterized in that the information is adjusted by inputting the position, height, radius, observation angle of the radar site. The method according to claim 6, The third step, The radar observation density analysis method of claim 1, wherein in displaying the lattice point of the observation radius, the observation radius is determined by specifying an angle for each radar site, and the altitude and radar altitude of the observation point are analyzed and calculated. The method according to claim 5, The analysis information of the five steps, Radar observation density analysis method characterized in that it is calculated through the comparative analysis of the quantitative precipitation estimation (QPE) reliability, correlation coefficient (R2), predictive error (RMSE) for a specific region. The method according to claim 5, The analysis information of the five steps, Radar observation density analysis method characterized in that the information to calculate the observation density and estimated radar precipitation performance (R ² ) for any area or observation blank using the quantitative precipitation estimation reliability and observation density relationship.

Images