KR20160068124A - Downburst detection system and method using the radar radial velocity - Google Patents

Abstract

Provided are a system and a method for detecting a downburst using a radar radial velocity. According to the present invention, the system is able to detect a downburst by detecting upper convergence which concurrently occurs with lower divergence, when the downburst is created, by using a radar elevation angle PPI radial velocity and reflectance data. Accordingly, a downburst detection accuracy rate of the present invention may be substantially higher than a downburst detection accuracy rate detected by only considering a lower divergence pattern.

Description

[0001] The present invention relates to a downburst detection system and method using a radar line velocity,

The present invention relates to a down-burst detection technique, and more particularly, to a radar line-of-sight detection method capable of improving down-burst detection accuracy by simultaneously considering a lower-layer divergence pattern and an upper- The present invention relates to a down-burst detection system and method.

The downburst is one of the unpredictable meteorological phenomena because strong wind shear occurs in a short span of time in a small area, but can be detected using the characteristics of the signal observed with the Doppler radar.

The downburst detection algorithm using Doppler radar has been mainly studied in the USA. However, the downburst algorithm using the Doppler weather radar network (the radar Doppler velocity) has been developed for the first time in Korea.

This algorithm detects the downburst by detecting the divergence pattern of the lower layer, which is one of the characteristics appearing in the field velocity field of the Doppler radar when the downburst occurs, and classifying it according to the reflectivity, sheer size, altitude and area threshold.

However, since only the lower divergence characteristic that occurs when a downburst occurs is used, there is a problem that the horizontal shear that appears in various weather phenomena is erroneously detected as a downburst.

In order to solve the above-described problems, the present invention provides a down burst detection method using a radar line speed that can improve a down burst detection accuracy by simultaneously considering an upper layer convergence pattern and a lower layer divergence pattern, And to provide a detection system and method.

That is, the present invention detects a horizontal divergence region of a downburst occurring within an altitude of 2 km by using a gaze velocity of a Doppler radar, detects a convergence region at an altitude of 2 km or more and detects a convergence region And determining a down burst in the presence of a matching area, thereby further improving the detection accuracy of the down burst.

According to another aspect of the present invention, there is provided a data collection unit for collecting observation data of Doppler weather radar, the observation data including radar PPI (Plan-Position Indicator) line speed data and radar PPI reflectivity data. -;

Using the collected lowest-layer radar PPI line-of-sight data, a divergence pattern of a line-of-sight velocity that varies from a negative line-of-sight velocity to a positive line-of-sight velocity is searched for as a searse segment and the detected searse segments are determined based on a preset threshold A divergence zone detection module for classifying the valid seer segments classified into divergence zones through grouping; Using the collected upper layer radar PPI line-of-sight velocity data, a convergence pattern of the line speed changing from the positive line-of-sight velocity to the negative line-of-sight velocity is found and detected as the line velocity convergence signal. Based on the detected threshold velocity convergence signals, A convergence region detection module for classifying the convergence region into a convergence region; And a down burst classification module for comparing whether the classified divergence region and the convergence region coincide with each other and classifying the region as a down burst if a convergence region of an upper layer coinciding with a divergence region of a lower layer as a comparison result exists, Speed downburst detection system.

According to another aspect of the present invention, there is provided a data collection method for collecting radar PPI gaze speed data and radar PPI reflectivity data; Detecting the divergence zone to detect the lower divergence zone using collected lowest radar PPI gaze velocity data; A convergence zone detection step of detecting the convergence region of the upper layer using the collected upper layer radar PPI line velocity data; And a downburst classification step of comparing the detected divergence region of the lower layer and the detected convergence region of the upper layer to classify the downburst, and a downburst detection method using the radar line velocity.

According to the present invention, it is possible to further improve the detection accuracy of the down burst by additionally detecting the upper layer convergence pattern together with the lower layer divergence pattern that appears when the down burst is generated from the observation data of the Doppler weather radar.

In addition, according to the present invention, by improving the detection accuracy of the down-burst, disaster caused by the down-burst can be prevented and the prevention factor of the airplane accident caused by the strong wind shear can be improved.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a schematic diagram illustrating a network of an overall system for down burst detection in accordance with an embodiment of the present invention.
2 is an exemplary diagram for explaining the divergence pattern of the down burst.
3 is an exemplary diagram showing the effect of the divergence pattern of the downburst on an aircraft landing.
4 is a block diagram showing a detailed configuration of the down burst detection system of FIG.
5 is an exemplary diagram for explaining a sheer segment classification function based on a threshold value of a predetermined gaze speed and reflectivity in a sheer segment detecting unit of the diverging region detecting module shown in Fig.
6 is an exemplary diagram for explaining the clustering function of the shear segment based on the azimuthal association window in the clustering unit of the diverging region detection module shown in FIG.
7 is an exemplary diagram for explaining a convergence signal classification function based on a threshold value of a line speed and a length set in advance in a line speed convergence signal classifying section of the convergence region detecting module shown in FIG.
Fig. 8 is an exemplary diagram for explaining a function for determining whether or not a divergence area and a convergence area coincide in the down burst determination module shown in Fig. 4; Fig.
9 is a flowchart for explaining a down burst detection method according to an embodiment of the present invention.
10 is a flowchart for explaining the sheer segment detection process in detail in the down burst detection method of FIG.
FIG. 11 is a diagram illustrating an example in which the detection result of the divergence zone detection process is compared with the divergence zone calculated by the AWS and the ground AWS observation data in the downburst detection method of FIG. 9; FIG.
FIG. 12 is a flowchart for explaining the process of detecting the convergence speed of the line-of-sight velocity in the down burst detection method of FIG.
FIG. 13 is a diagram showing an example of a detection result obtained by applying the gaze speed convergence signal detection process of FIG. 12 to an actual case.
FIG. 14 is an exemplary diagram showing an upper layer convergence region coinciding with a lower layer divergence region by applying a down-burst determination process in the down-burst detection method of FIG. 9;

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Reference will now be made in detail to the preferred embodiments of the present invention, examples of which are illustrated in the accompanying drawings. In the following description, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear.

In the present invention, a gaze divergence pattern of the down-burst is defined as a sheer segment in order to detect a down-burst. That is, when a downburst occurs, a descending wind hits the ground and spreads in all directions, and a divergence pattern that changes from a negative gaze velocity to a positive gaze velocity appears in the radar line velocity field. The gaze velocity pattern is defined as a shear segment.

In the present invention, the line-speed convergence pattern of the down-burst is defined as the line-of-sight convergence signal. That is, when a downburst occurs, a convergence pattern that changes from a positive line-of-sight velocity to a negative line-of-sight velocity appears in the radar line-of-sight velocity field of the upper layer. do.

The present invention will be described with reference to these definitions.

First, a down-burst detection system according to an embodiment of the present invention will be described with reference to Figs. 1 to 3. Fig.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a schematic diagram illustrating a network of an overall system for down burst detection in accordance with an embodiment of the present invention. 2 is an exemplary diagram for explaining the divergence pattern of the down burst. 3 is an exemplary diagram showing the effect of the divergence pattern of the downburst on an aircraft landing.

Referring to FIG. 1, an overall system using a down-burst detection system according to the present invention may include at least a Doppler meteoroid radar system 10 and a down-burst detection system 20.

The Doppler weather radar system 10 observes the weather at a constant location and transmits the observed data to the down burst detection system 20 wirelessly or by wire. Here, the observed data include radar PPI (Plan-Position Indicator) reflectivity data and radar PPI line speed data.

For reference, the Doppler weather radar is a radar that applies the Doppler effect in which the reflected wave frequency fluctuates according to the velocity of the target when a radio wave hits a moving target. Therefore, And is an important observation instrument for safe navigation of aircraft.

Since the gaze velocity calculated from the Doppler weather radar is far from the radar direction or represents the velocity component of the approaching object, it is suitable for detecting the downburst which locally changes the wind direction and wind speed locally.

In this embodiment, the Doppler meteorological radar system 10 is composed of eleven operational radars currently operated by the Korea Meteorological Administration, and meteorological observation data on the Korean region can be obtained through this.

Then, the downburst detection system 20 detects the lower divergence region and the upper-layer convergence region based on the observation data collected from the Doppler meteorological radar system 10 and compares the detected lower divergence region and the upper-layer convergence region, It is possible to detect the down-burst. Here, the lower divergence region can be detected using a shear segment that defines a divergence pattern.

Here, as shown in Figs. 2 (a) to 2 (c), the downburst is a blast due to a strong downward airflow, and when a downburst occurs, the downward wind bumps against the ground and spreads in all directions, A divergence pattern that changes from a negative gaze velocity to a positive gaze velocity in a radar line-of-sight velocity field is generated, as shown in (d) of FIG.

This diverging pattern of the downburst makes it possible to prevent a strong downward force that forces the aircraft downward while moving horizontally when the aircraft passes the downburst, At this time, the aircraft will be downward toward the ground. If the intensity of the downward gusts is high, collision with the ground may occur.

Accordingly, the down-burst detection system 20 according to the present invention detects a shear segment that defines a divergence pattern of a down-burst, finds a lower divergence region, and detects a convergence signal of a gaze velocity to find an upper- By comparison, it is possible to finally detect the down-burst.

This downburst detection system 20 collects the gaze velocity data and reflectivity data of the PPI generated in the Doppler weather radar system 10, that is, 11 operating radar networks currently operating in the weather station, After performing the management, sheer segments are detected from the line velocity and reflectivity data of the lowest layer PPI, and the detected sheer segments are classified based on the length, altitude, line speed, and reflectivity threshold.

Then, the downburst detection system 20 groups clusters based on the distribution density of azimuthal angles and classifies the clusters into lower divergence regions based on the area threshold value. Details of the detection process of the lower layer divergence region are described below.

Also, the down-burst detection system 20 detects the line-speed convergence signal from the line-speed and reflectivity data of the upper layer PPI, and classifies the detected line-speed convergence signal into the upper-layer convergence region based on the threshold value. The detailed upper-layer convergence region detection process will be described in detail below.

As described above, the down-burst detection system according to the present invention is suitable for the Korean environment by detecting the down-burst using observation data of eleven operational radar networks installed and operated throughout Korea, By using the convergence signal together, a more accurate down-burst can be detected.

Next, the detailed configuration of the down-burst detection system will be described with reference to Figs. 4 to 8. Fig.

4 is a block diagram showing a detailed configuration of the down burst detection system of FIG. 5 is an exemplary diagram for explaining a sheer segment classification function based on a threshold value of a predetermined gaze speed and reflectivity in a sheer segment detecting unit of the diverging region detecting module shown in Fig. 6 is an exemplary diagram for explaining the clustering function of the shear segment based on the azimuthal association window in the clustering unit of the diverging region detection module shown in FIG. 7 is an exemplary diagram for explaining a convergence signal classification function based on a threshold value of a line speed and a length set in advance in a line speed convergence signal classifying section of the convergence region detecting module shown in FIG. Fig. 8 is an exemplary diagram for explaining a function for determining whether or not a divergence area and a convergence area coincide in the down burst determination module shown in Fig. 4; Fig.

4, the down-burst detection system 20 includes at least a data collection unit 21, a diverging area detection module 22, a convergence zone detection module 23, and a down-burst classification module 24 .

First, the data collection unit 21 collects observation data from the Doppler meteorological radar system 10. Here, the collected observation data are the reflectivity data and gaze speed data of the weather-radar PPI (Plan-Position Indicator).

In addition, the data collection unit 21 performs quality management of the collected observation data. That is, the data collecting unit 21 may perform noise reduction on observation data using, for example, a 3 × 3 average filter to reduce the influence of observation errors of the visual speed.

In addition, the data collection unit 21 extracts data at a point 5 km away from the Doppler weather radar from observational data in order to prevent the shear segment detection error due to the sight line discontinuity point caused by the terrain in the vicinity of the Doppler weather radar You can do the work.

The diverging area detecting module 22 detects the diverging area from the observation data transmitted from the data collecting part 21 by using the sheer segment detecting part 221, the clustering part 222 and the divergence area classifying part 223 ). ≪ / RTI >

The sheer segment detection unit 221 compares the two gaze velocities at arbitrary azimuth angles from the lowermost radar PPI gaze velocity data among the collected observations and searches for a point where the gaze velocity increases as a result of comparison, Designates the start point of the segment, searches for a point at which the gaze speed decreases as a result of the comparison, and specifies the end point of the shear segment.

At this time, in order to reduce the influence of the error data at the start point search of the shear segment, the sheer segment detecting unit 221 sets the two gaze velocity differences to a point of 5 ms ^-1 or less and also searches for the end point of the shear segment We use the sum_decrease variable, which indicates the tendency of the decrease in the visual velocity.

In other words, the sheer segment detecting unit 221 initializes the sum_decrease variable to '0' at a specified starting point after designating a point at which the line speed is increased as a starting point, and when a decrease in line speed is detected Set the maximum line speed (max_radial velocity) at the point where the reduction starts, add the sum_decrease variable to the sum_decrease variable by the difference in line speed, and designate the point exceeding the set threshold value (5 ms ^-1 ) as the end point of the searse segment do. If the increased gaze speed is again increased, the increased gaze speed is compared with the maximum gaze speed. If the increased gaze speed is less than the maximum gaze speed, then the gaze speed difference is added to the sum_decrease variable, Continue the search for the point and decrease the sum_decrease variable by 1 ms ^-1 if the increased gaze speed is greater than the maximum gaze speed. Also, if the tendency of increasing eye velocity continues, the sum_decrease variable is decreased by 1 ms ^-1 and initialized to '0' at the time of negative value. At this time, in this embodiment, when the increased gaze speed is greater than the maximum gaze speed, the decrease width of the sum_decrease variable is set to be constant at 1 ms ^-1 because the influence of the large increase / decrease in the gaze speed discontinuity point due to the error data And to give a greater weight to detecting the decreasing tendency.

Through this process, the sheer segment detection unit 221 detects the sheer segment through the starting point and the end point of the detected sheer segment, and transmits the detected theoretic segment information to the clustering unit 222.

Then, the clustering unit 222 performs an operation of classifying the found theoretical segments based on a preset threshold value. Here, the predetermined threshold value is a value set based on the definition of the divergence pattern of the down burst, and includes a length, a shear, an altitude, a reflectivity, etc. For example, the difference between the maximum value and the minimum value of the line- The sheer shall be not less than 10 ms ^-1 , the reflectivity ratio of not less than 35 dBZ shall be not less than 50%, the length shall be not less than 2 km, the altitude shall be not more than 2 km, In order to detect the change in velocity, the gaze segment must include a gaze velocity of 0 ms ^-1 .

In other words, the clustering unit 222 classifies the corresponding theearsegments by applying a predetermined threshold value to the founder theearse segments. That is, the clustering unit 222 can check whether the gaze speed of 0 ms ^-1 is included in each sheer segment, and sort the sheer segments based on the confirmation result.

Also, the clustering unit 222 can check whether the length of the sheer segment is 2 km or more and the altitude is 2 km or less, and classify the sheer segments based on the confirmation result.

5A, the clustering unit 222 checks whether the shear indicating the difference between the maximum value and the minimum value of the gaze speed in each classified sheer segment is 10 ms ^-1 or more, Based on the results, sheer segments can be classified. That is, if the difference between the maximum value and the minimum value of the gaze speed is 10 ms ^-1 or more, it is classified into the effective seam segment.

5 (b), the clustering unit 222 checks whether the ratio of the area where the reflectivity of the shear segments is 35 dBZ or more is 50% or more, classifies the theear segments based on the confirmation result . That is, if the ratio of the area having the reflectivity of 35 dBZ or more is 50% or more, it is classified as an effective seam segment.

Here, the application order of the threshold values is arbitrary, and if all the threshold values are satisfied, they are eventually classified into the valid segment.

Through the sheer segment classification operation, the clustering unit 222 can classify the valid theoretic segments and the non-valid theoretic segments.

Then, the clustering unit 222 performs clustering using the azimuthal association window with respect to the sorted effective valid seed segments.

That is, the clustering unit 222 constructs clusters by grouping the effective weir segments adjacent to each other at arbitrary azimuth angles into one group using an azimuthal association window. The effective weir segment included in the window Check that 50% or more of the valid shear segments exist in the window. If the result is more than 50%, the valid shear segments in the window are grouped into one group to form a cluster. This clustering process occurs until a window with a sheer segment of less than 50% occurs, and the sheer segment group, or cluster, grows.

In this embodiment, as shown in FIG. 6, considering that the observation radius increases as the distance increases from one azimuth angle, the azimuthal association window is referred to as a radar site (a point at 0 km) 11 windows are set, and as the distance increases, the number of windows is gradually decreased, so that three windows can be set at a point 10 km or more from the radar site.

Then, the divergence area classifier 223 classifies clusters, which are groups of valid theoretical segments, as divergence regions, based on the area threshold value.

That is, the divergence area classifying unit 223 classifies the clusters into divergence areas when the area of the clusters is obtained, and the area of the clusters obtained by comparing the areas of the obtained clusters with preset area threshold values is equal to or greater than the area threshold value. At this time, the area threshold value for classifying the divergence area is a value obtained by the following equation (1).

Here, when A _min = 0.5 km, R _slope = 0.1 km ² km ^-1 , R _min = 2 km, R _max = 16 km, Shear _min = 10.0 ms ^-1 and Shear _max = 27.5 ms ^-1 ,

이고,

ego,

이다.

to be.

In other words, the divergence area classifier 223 calculates the area threshold value using Equation 1, which is a function of the length R and Shear, wherein the length R is at least 2 km, The actual sheer segment cluster length is used and if the Shear is 10.0 ms ^-1 or more and 27.5 ms ^-1 or less, the sheer value of the actual sheer segment cluster is used. Then, the divergence area classifier 223 classifies the divergence area for the cluster using the obtained area threshold value.

The convergence area detection module 23 detects the convergence area from the observation data transmitted from the data collection unit 21 by using the gaze speed convergence signal detection unit 231 and the gaze speed convergence signal classification unit 232 At least.

The line-of-sight convergence signal detecting unit 231 compares two line-of-sight velocities at arbitrary azimuth angles from the upper-layer radar PPI line-of-sight velocity data among observation data transmitted from the data collecting unit 21, The point at which the speed decreases is designated as the start point of the glaze, and the point at which the gaze speed increases is designated as the end point of the sheer segment. At this time, the line-of-sight convergence signal detecting unit 231 uses a sum_increase variable indicating an increasing trend of the line speed to search for an end point of convergence.

In other words, the line-of-sight convergence signal detecting unit 231 initializes the sum_increase variable to '0' at the designated starting point after designating the point where the line speed decreases, as the starting point, If it is detected, the line speed at the point where the increase starts is set as the minimum line speed (min_radial velocity), and the point exceeding the set threshold value (3 ms ^-1 ) added to the sum_increase variable by the line speed difference is set as the end point of convergence Specify. If the decreased gaze speed is smaller than the minimum gaze speed, the sum of the gaze speeds is added to the sum_increase variable, and the convergence end Continue the search for the point, and decrease the sum_increase variable by 1 ms ^-1 if the decreased gaze speed is less than the minimum gaze speed. If the tendency of decreasing visual velocity continues, the sum_increase variable is decreased by 1 ms ^-1 and initialized to '0' at the time of negative value. At this time, in the present embodiment, when the decreased visual velocity is smaller than the minimum visual velocity, the decrease width of the sum_increase variable is set to be constant at 1 ms ^-1 because the influence of the large increase / decrease at the visual velocity discontinuity point due to the error data And to give a greater weight to detecting increasing trends.

Through this process, the line-of-sight convergence signal detecting unit 231 detects the line-of-sight convergence signal through the start point and the end point of the found convergence and outputs the detected line-of-sight convergence signal information to the line- (232).

The line-of-sight convergence signal classifying unit 232 classifies the retrieved line-of-sight velocity convergence signals into a convergence region based on a preset threshold value. Here, the preset threshold value is set based on the definition of the convergence pattern of the down burst, and may include the line speed, the length, the altitude, the reflectivity, and the like. For example, in the present embodiment, in order to detect a change from a positive gaze velocity to a negative gaze velocity, a gaze velocity of 0 ms ^-1 should be included in the gaze velocity convergence signal section, and a length of 2 km or more , And the convergence altitude is set to 2 km or more in consideration of the fact that the detection altitude of the lower layer divergence region is set to 2 km or less, and the reflectance ratio of the reflectivity of 35 dBZ or more may be set to include 50% or more. In the case of the reflectivity, the position of the upper-layer convergence signal according to the down-burst generation mechanism appears in a place where the reflection core descends, near the melting point altitude, and near the low cloudiness (Roberts and Wilson, 1989) Is set to include at least 50% of the reflectance ratio of 35 dBZ or more.

In other words, the line-of-sight convergence signal classifying unit 232 classifies the corresponding line-of-sight velocity convergence signals by applying predetermined threshold values to the detected line-of-sight velocity convergence signals.

That is, as shown in Fig. 7, the gaze speed convergence signal classifying section 232 determines whether or not the gaze speed of 0 ms ^-1 is included in each gaze speed convergence signal, the length of the gaze speed convergence signal is 2 km or more The altitude is 2 km or more, and the gaze velocity convergence signals can be classified based on the confirmation result.

Also, the line-of-sight convergence signal classifying unit 232 may classify the line-of-sight velocity convergence signals based on the result of checking whether the ratio of the area where the reflectance of the line-velocity convergence signals is 35 dBZ or more is 50% or more. That is, if the ratio of the region having the reflectivity of 35 dBZ or more is 50% or more, it can be classified as the convergence region. Here, the application order of the threshold values is arbitrary, and if all the threshold values are satisfied, the convergence region is finally classified.

Through the classification operation of the line-speed convergence signal, the line-of-sight convergence signal classifying unit 232 can classify the convergence region from the line-speed convergence signals.

Then, the down burst classification module 24 performs a function of detecting a converging region existing in an upper layer of the diverging region detected in the lower layer.

First, the down burst classification module 24 determines a range of 1 km in the divergence area detected by the divergence area detection module 22, and removes the convergence area of the upper layer existing outside the divergence area. At this time, the range for classifying the convergence region is broadened to 1 km in order to consider the volume scan time difference for each altitude angle.

Thereafter, the down burst classification module 24, by comparing the lower layer divergence area and the upper layer convergence area, leaves only the upper layer convergence area coinciding with the lower layer divergence area, and finally classifies the area as the down burst.

That is, as shown in Fig. 8, the down burst classification module 24 can classify a down burst into a region where the divergence region of the lower layer coincides with the convergence region of the upper layer.

Next, the down burst detection method according to the present invention will be described with reference to FIGS. 9 to 14. FIG. Here, the present down-burst detection method can be implemented by the down-burst detection system described with reference to Figs.

9 is a flowchart for explaining a down burst detection method according to an embodiment of the present invention. 10 is a flowchart for explaining the sheer segment detection process in detail in the down burst detection method of FIG. FIG. 11 is a diagram illustrating an example in which the detection result of the divergence zone detection process is compared with the divergence zone calculated by the AWS and the ground AWS observation data in the downburst detection method of FIG. 9; FIG. FIG. 12 is a flowchart for explaining the process of detecting the convergence speed of the line-of-sight velocity in the down burst detection method of FIG. FIG. 13 is a diagram showing an example of a detection result obtained by applying the gaze speed convergence signal detection process of FIG. 12 to an actual case. FIG. 14 is an exemplary diagram showing an upper layer convergence region coinciding with a lower layer divergence region by applying a down-burst determination process in the down-burst detection method of FIG. 9;

Referring to FIG. 9, a method of detecting a downburst according to the present invention includes a process of performing observation data collection and quality control (S301), a divergence zone detection process (S302), a convergence zone detection process (S303) (S304).

The observation data collection and quality control process (S301) collects line-of-sight velocity data and reflectivity data of the Plan-Position Indicator (PPI) generated from eleven operational radar networks and uses a 3 × 3 average filter And performing a quality control process.

Then, a divergence zone detection process S302 is performed. The divergence zone detection process S302 includes a sheer segment detection process S312, a sheer segment classification process S322, a sheer segment clustering process S332, And a classification process (S342).

Here, the sheer segment detection process (S312) is a process of searching for a start point and an end point of the shear segment. The two gaze velocities at one azimuth angle are compared with each other to search for an increasing point and designated as a start point of the shear segment And finds the end point of the shear segment using the sum_decrease variable indicating the tendency of the decrease in the gaze speed. In the present embodiment, the starting point of the sheer segment is set to a point where the two line speed differences are equal to or less than 5 ms ^{-1 in} order to reduce the influence of the error data.

More specifically, referring to FIG. 10, the sheer segment detection process (S312) will be described. The two gaze velocities V _i and V _{i + 1} are compared at one azimuth angle to search for a point at which the gaze velocity increases (S401), and when the point at which the line speed is increased is searched (S401, Y), it is designated as the start point. At this time, the sum_decrease variable (SD _i ) is initialized to '0 ms ^-1 ' at the start point.

Then, when the reduction of the radial velocity detection (S401, N), the radial velocity of a point at which the reduction starts, the maximum radial velocity; is set to (V _M max_radial velocity) as the difference between the radial velocity (V _D) sum_decrease variables ( added to the SD _i) (S407, S408) the threshold (5 ms ^-1) point (S409, Y exceeds the set) is the end point of the shear segments (S410).

If the tendency of the decrease of the gaze speed is increased again (S409, N, S401, Y), the value of the gaze speed V _{i + 1} is compared with the maximum gaze speed V _M (S402) If the velocity value V _{i + 1} is smaller than the maximum gaze velocity V _M (S402, N), the gaze velocity difference V _D is added to the sum_decrease variable SD _i (S407) It is checked if the sum_decrease variable (SDi _-1 ) prior to (S402, Y) is 0 ms ^-1 if the increased gazing velocity value V _{i + 1} is larger than the maximum gazing velocity V _M If it is not to (S403) reduces the sum_decrease variable (SD _i) as much as 1 ms ^-1 (S405).

And if the tendency of increasing radial velocity continued to reduce the sum_decrease variable (SD _i) as long as 1 m / s and initializes the variables sum_decrease (SD _i) at the time the negative (S406) to '0' (S404).

The reason why the decrement of the sum_decrease variable (SD _i ) is kept constant at 1 ms ^-1 when the increased visual velocity value (V _{i + 1} ) is larger than the maximum visual velocity (V _M ) In order to reduce the impact of large increases and decreases in trends.

Then, a sheer segment classification process (S322) is performed. The process (S322) is a process of classifying the discovered sheer segments by applying segment threshold values relating to the length, altitude, to be. In this case, the application of the segment threshold value for classifying the sheer segment may apply all segment threshold values according to the sheer segment or may classify the sequential shear segments according to the segment threshold values, and the application order of the segment threshold values may be arbitrary.

For example, it is possible to determine whether the difference between the maximum value and the minimum value of the gaze speed is 10 ms ^-1 or more, whether the gaze speed is 0, the length is 2 km or more, the altitude is 2 km or less, Seed segments that satisfy pre-established segment threshold conditions are classified, such as if the ratio of the reflectivity region over 35 dBZ is 50% or more. Here, the reason that the sheer segment should include the line speed of '0 ms ^-1 ' is to detect the change from the negative line speed to the positive line speed, and the height of the sheer segment, considering the altitude of the radar sites , And 2 km or less.

Then, a sheer segment clustering process (S332) is performed. In this process (S332), sheer segments neighboring each other for an arbitrary azimuth angle are grouped into one group using an azimuthal association window This is the process of creating a cluster.

Specifically, using the azimuthal association window, it is determined whether the sheer segment existing in the neighboring radius is a valid sheer segment, that is, whether the two sheer segments are sufficiently close on the azimuth angle. Then, If the segment is a valid thesegment, it is bundled into one cluster. At this time, the distribution density of the sheer segment is used as a threshold value, and it can be set to be effective when the distribution density of the sheer segment is 50% or more.

Meanwhile, as shown in FIG. 6, it is preferable to set the number of windows of the azimuthal association window to be different considering that the observation radius increases as the distance increases from one azimuth angle. In this embodiment, the number of windows included in the azimuthal association window is set to 11 in the radar site, and as the distance increases, the number of windows is gradually decreased to be 3 Set to.

That is, if there are more than 50% of the neighboring theater segments in the set number of windows, the corresponding theater segments are grouped into one group to form a cluster, and the neighboring theater segments are set to 50 %, And if there is more than 50%, the process of grouping the corresponding thether segments into one group is repeated until there are less than 50% of the thether segments in the window. Through this grouping process, the clusters for the sheer segment group become larger.

Then, a cluster classification process S342 is performed. In this process S3442, clusters are classified into a divergence region using a preset area threshold value, and the divergence pattern of actual downburst It is a task to remove clusters that do not cause the phenomenon.

That is, clusters having a predetermined area threshold value (A _th ) or more among clusters configured in the theoretical segment clustering process (S332) are classified as divergence regions of the down burst. In this case, the area threshold value A _th is a function of the length R and the shear, and the length Rr of the sheer segment group is 2 km or more and 16 km or less , The length of the actual cluster is used. If the range is not satisfied, the minimum length value _Rmin or the maximum length value _Rmax is used. Further, the shear in the cluster (Shear _r) is 10.0 ms ^-1 or more, yet is less than 27.5 ms ^-1 when using the shear value of the physical clusters, and does not meet the minimum range, the shear (Shear _min) or the maximum shear ( Shear _max ).

11, the divergence area detected through the divergence zone detection process (S302) is verified through comparison with the time series data of the divergence field and the AWS (Automatic Weather System) do.

11 (a) shows an image of a divergence field detected at 21:50 on July 12, 2012 in Kunsan area as a divergence field, and FIG. 11 (b) shows a wind vector data of a ground AWS. Here, the data used in the present invention is a case of a single convection cell down-burst observed in Gunsan area, 50 minutes on July 12,

11A and 11B, the downburst region coincides with the reflectivity of 50 dBZ or more, and the change in the line-of-sight velocity coincides with -5 ms ^-1 in the diverging region (indicated by a black circle in (a) It can be seen that this is also consistent with the area observed in the divergence field calculated by AWS (the area indicated by the black arrow in (b)).

Then, a convergence region detection process S303 is performed. The convergence region detection process S303 may include a gaze-speed convergence signal detection process S313 and a gaze-speed convergence signal classification process S323.

Here, the gaze speed convergence signal detection process S313 is a process of finding a starting point and an end point of the gaze speed convergence signal. The gaze speed convergence signal detection process S313 searches for a point where the two gaze velocities are compared and decreased at one azimuth angle, And the end point of the convergence speed of the gaze speed is searched by using the sum_increase variable indicating the increasing trend of the gaze speed. In this embodiment, the starting point of the line-speed convergence signal is set to a point where the difference in the line-of-sight velocities is 3 ms ^-1 or less in order to reduce the influence of the error data.

More specifically, referring to FIG. 12, the gaze speed convergence signal detection process (S313) will be described. The gaze speed convergence signal detection process (S313) will be described. The gaze speed convergence signal detection process Is designated as the start point.

Thus, the radial velocity convergence signal detection process of the present invention (S313) is made that the start point are detected on the assumption, sum_increase variable _{(SI i-1, i =} 2) at the start point and the minimum radial velocity VR _min is 0 ms ^-1 (S501).

Then, when an increase in the gaze speed is detected (S502, N), the gaze speed VR _i-1 at the point where the increase is started is set to the minimum gaze speed VR _min (S507, Y) by car (VR_D) of the variable speed sum_increase added to the _{(SI i) (S508, S509} ) the threshold set (3 ms ^-1) than the point (S510, Y) are detected the end point of the converging radial velocity signal (S511 ).

If the trend of who increased radial velocity is reduced again, the value of the reduced radial velocity (VR _i) the minimum radial velocity (VR _min) of the radial velocity down to a comparison of the value value (VR _i) the minimum radial velocity (VR _min , the difference in line speed VR_D is added to the sum_increase variable SI _i in step S509 to continuously search for the end point of the line speed convergence signal. However, if the reduced line speed value VR _i is less than the minimum line speed VR _min ), if the previous sum_increase variable SI _i-1 is not 0 ms ^-1 (S 503, N), the sum_increase variable SI _i ^-1 is decreased by 1 ms ^-1 S505).

If the decrease of the gaze speed continues, the sum_increase variable SIi _-1 is reduced by 1 ms ^-1 and the sum_increase SI _i-1 is initialized to 0 at the time of the negative value S506, Y S504).

The reason why the decrease width of the sum_increase variable (SI _i-1 ) is kept constant at 1 ms ^-1 when the reduced visual velocity value VR _i is smaller than the minimum visual velocity VR _min is that the visual velocity discontinuity point In order to reduce the impact of large increases and to give greater weight to detecting increasing trends.

The result of applying the detection process S313 of the line-of-sight velocity convergence signal to the case of the down-burst is illustrated in FIG.

13, the blue solid line indicates the gaze speed, the red bar indicates the difference in the gaze speed, the gray solid line indicates the sum_increases, and the green triangle indicates the start and end points, where 1 is the start point and 3 is the end point At BIN 141, the flag starts at 1 and starts to increase. At BIN 161, the sum_increases variable is greater than 3, and the signal is terminated.

Subsequently, a classification process S323 of the line-of-sight velocity convergence signal is performed. In this process S323, the line-of-sight velocity convergence signal is calculated by applying threshold values relating to the length, altitude, It is a process of classifying signals.

For example, if the seek velocity convergence signals have a length of 2 km or more, an altitude of 2 km or more, a line speed of '0', a ratio of a reflectivity region of 35 dBZ or more is 50% , And classifies the line-of-sight velocity convergence signals satisfying by applying preset threshold conditions. Here, the reason why the line speed convergence signal should include the line speed of '0 ms ^-1 ' is to detect the change from the positive line speed to the negative line speed.

Next, a downburst detection process S304 is performed. The downburst detection process S304 is a process of classifying a convergence region existing in an upper layer of the divergence region detected in the lower layer through a divergence zone detection process S302 .

That is, the downburst sorting process S304 may be performed by comparing the divergence region of the lower layer detected through the divergence zone detection process S302 and the convergence region of the upper layer detected through the convergence zone detection process S303, And the corresponding region can be finally classified as a down-burst.

The result of applying the down burst classification process (S304) to the down burst case will be illustrated in FIG.

Fig. 14 (a) shows the PPI image of the divergence area detected at the lowest layer observed in Kunsan radar, Fig. 14 (b) shows the convergence area PPI image detected at the line velocity field of the seventh altitude PPI, And PPI images of the upper layer convergence region.

Referring to FIGS. 14A to 14C, the convergence regions detected in the upper layer are classified according to the threshold values, and appear in a total of 4 places, and the largest convergence region coincides with the divergence region shown in the lower layer, Bursts.

As described above, according to the down-burst detection method of the present invention, it is possible to detect the accurate down-burst by detecting both the divergence region of the lower layer and the convergence region of the upper layer, which occurs when the down- burst occurs, based on the mechanism of the down-burst.

The present invention has been described with reference to the preferred embodiments. It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. Accordingly, the disclosed embodiments are to be considered in an illustrative rather than a restrictive sense, and all differences within the scope of equivalents thereof should be construed as being included in the present invention.

Further, the apparatus and method according to the present invention can be embodied as computer readable codes on a computer-readable recording medium. A computer-readable recording medium includes all kinds of recording apparatuses in which data that can be read by a computer system is stored. Examples of the recording medium include a ROM, a RAM, a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like, and a carrier wave (for example, transmission via the Internet). The computer-readable recording medium may also be distributed over a networked computer system so that computer readable code can be stored and executed in a distributed manner.

10. Doppler weather radar system 20. Downburst detection system
21. Data collection module 22. Divergence zone detection module
23. convergence zone detection module 24. down burst classification module

Claims (13)

A data collection unit for collecting Doppler weather radar observations; the observations include radar PPI (Plan-Position Indicator) line-of-sight data and radar PPI reflectance data;
Using the collected lowest-layer radar PPI line-of-sight data, a divergence pattern of a line-of-sight velocity that varies from a negative line-of-sight velocity to a positive line-of-sight velocity is searched for as a searse segment and the detected searse segments are determined based on a preset threshold A divergence zone detection module for classifying the valid seer segments classified into divergence zones through grouping;
Using the collected upper layer radar PPI line-of-sight velocity data, a convergence pattern of the line speed changing from the positive line-of-sight velocity to the negative line-of-sight velocity is found and detected as the line velocity convergence signal. Based on the detected threshold velocity convergence signals, A convergence region detection module for classifying the convergence region into a convergence region; And
A down burst classification module that compares the classified divergence divergence region with a convergence region and classifies the region as a down burst if a convergence region of an upper layer coinciding with a divergence region of a lower layer as a result of the comparison is present;
And a radar line velocity including the radar line velocity. The method according to claim 1,
The convergence zone detection module finds the start point and end point of the convergence pattern of the line speed changing from the positive line speed to the negative line speed using the collected upper layer radar PPI line speed data, A line velocity convergence signal detecting unit for detecting an area formed by the line velocity convergence signal; And
A line-of-sight convergence signal classifying unit for comparing the detected line-of-sight velocity convergence signals with a threshold value including at least length, altitude, line speed and reflectivity, and classifying the convergence region based on the comparison result;
A radar line velocity detection unit for detecting a radar line velocity; The method of claim 2,
The line-of-sight convergence signal detecting unit compares two line-of-sight velocities for one azimuth angle, designates a point where the line-of-sight velocity decreases from the comparison result as a starting point of the line-of-sight velocity convergence signal, Wherein a point where the sum_increase variable indicating the degree of increase of the line speed is greater than a preset threshold value is designated as the end point. The method of claim 3,
The gaze speed convergence signal detecting unit initializes the sum_increase variable to 0 at a specified starting point and sets the gaze speed at the point where the gaze speed increase starts when the gaze speed increase is detected to the minimum gaze speed, The sum of the gaze speeds is added to the sum_increase variable, and if the sum is greater than or equal to a predetermined threshold value, the point is designated as the end point. The method of claim 2,
The line-of-sight convergence signal classifying unit classifies the detected line-of-sight convergence signals whether the length is greater than or equal to 2 km, whether the altitude is greater than or equal to 2 km, whether the line speed includes 0 ms ^-1 , Area rate is 50% or more, and all the satisfying convergence speed convergence signals are classified into a convergence area. The method according to any one of claims 1 to 5,
The diverging zone detection module finds the starting point and ending point of the divergence pattern of the line velocity changing from the negative gaze velocity to the positive gaze velocity using the collected lowest-layer radar PPI gaze velocity data, A sheer segment detecting unit for detecting a region formed by the shear segment;
The length, shear, altitude, and reflectivity of the detected thether segments are compared with preset threshold values, classified into valid theoretic segments based on the comparison result, and the distribution of the predefined azimuth aggregate windows A clustering unit for constructing a cluster by grouping valid seed segments satisfying a density threshold value; And
And a divergence area classifier for comparing the configured clusters with a preset area threshold value and classifying the constructed clusters into a divergence area based on the comparison result. A data collecting step of collecting radar PPI line-of-sight data and radar PPI reflectance data;
Detecting the divergence zone to detect the lower divergence zone using collected lowest radar PPI gaze velocity data;
A convergence zone detection step of detecting the convergence region of the upper layer using the collected upper layer radar PPI line velocity data; And
A down burst classification step of comparing the detected divergence region of the lower layer and the detected convergence region of the upper layer to classify the down burst;
And detecting a downward burst using the radar line velocity. The method of claim 7,
The diverging region detection step finds a starting point and an ending point of a divergence pattern of a line velocity changing from a negative gaze velocity to a positive gaze velocity using the collected lowest-order radar PPI gaze velocity data, A sheer segment detection step of detecting a region formed by the sheer segment;
A valid weir segment classification step of classifying valid weir segments based on a preset threshold value from detected heuristic segments;
A cluster constructing step of constructing a cluster based on a threshold value of a predetermined distribution density of the sorted effective segments; And
And a divergence region classification step of classifying the configured clusters into divergence regions based on a preset area threshold value. The method of claim 7,
The convergence region detection step uses the collected upper layer radar PPI line-of-sight data to find the start point and end point of the convergence pattern of the line speed changing from the positive line speed to the negative line speed, A line speed convergence signal detecting step of detecting an area made up of the line speed convergence signal; And
A line-of-sight convergence signal classifying step of classifying convergence regions based on a threshold value including at least the detected line-velocity convergence signals at a length, an altitude, a line-of-sight velocity, and a reflectivity;
Wherein the radar line velocity is calculated based on the radar line velocity. The method of claim 9,
The convergence signal detection step may include comparing two neighboring line-of-sight velocities for one azimuth using the collected upper-layer radar PPI line-of-sight velocity data;
If it is judged that the gaze speed is decreased, the point where the gaze speed decreases is designated as the starting point of the gaze speed convergence signal, and the sum_increase variable indicating the degree of increase of the gaze speed is initialized to 0 and the gaze speed Setting a gaze speed;
Comparing a value obtained by adding the difference between two line speeds to the sum_increase variable and a predetermined threshold value when an increase in line speed is detected as a result of comparison between two neighboring line speeds after a designated start point; And
Designating a corresponding point as an end point if the sum_increase variable plus a difference in two line speeds is greater than the preset threshold value;
Wherein the down burst detection method comprises the steps of: The method of claim 10,
Wherein the step of detecting the line-of-sight convergence signal comprises the steps of: comparing a reduced line-of-sight velocity value and a minimum line-of-sight velocity value when a decrease in the increased line-of-sight velocity is detected;
If the decreased gaze velocity value is greater than the minimum gaze velocity, adding the difference of the two gaze velocities to the sum_increase variable and continuing to search for the end point of the gaze velocity convergence signal; And
Decreasing the sum_increase variable by 1 ms ^-1 and initializing the sum_increase variable to zero when the value of the sum_increase variable becomes negative if the gaze speed value decreased as a result of comparison is smaller than the minimum gaze speed;
And detecting the radar line velocity based on the radar line velocity. The method of claim 9,
The line-of-sight convergence signal classifying step may include classifying the detected line-of-sight convergence signals as to whether the length is greater than or equal to 2 km, whether the generated altitude is greater than or equal to 2 km, whether the line velocity includes 0 ms ^-1 , determining whether a ratio of dBZ or more is 50% or more; And
Classifying the line-of-sight convergence signal having a length of 2 km or more, a generation altitude of 2 km or more, a line speed of 0 ms ^-1 and a reflectivity of 35 dBZ or more as 50% or more as a convergence area;
Wherein the down burst detection method comprises the steps of: 13. A recording medium on which a program readable and executable by a digital signal processing apparatus for performing the method according to any one of claims 7 to 12 is recorded.

Images