KR101221755B1 - Method for identifying reflectivity cells associated with severe weather - Google Patents

Abstract

The present invention utilizes three-dimensional radar reflectivity fields to detect reflective cells associated with dangerous weather that can accurately detect reflective cells associated with dangerous weather such as thunderstorms, hail and heavy rain. A method, comprising: a first step of progressively identifying one-dimensional, two-dimensional, and three-dimensional components in a grid point component of 3D CAPPI data to detect reflectivity cells associated with dangerous weather; And a second step of calculating an attribute of the reflectivity cell associated with the detected dangerous weather.

Description

Method for identifying reflectivity cells associated with severe weather

FIELD OF THE INVENTION The present invention relates to a dangerous weather detection method, and more particularly, to reflectance cells associated with dangerous weather such as thunderstorm, hail and heavy rain using three-dimensional radar reflectivity fields. The present invention relates to a detection method of a reflectance cell associated with a dangerous meteorology capable of accurately detecting.

In general, meteorological radar is a device that calculates the magnitude of radio signals reflected or scattered by meteorological targets by emitting electromagnetic waves, and monitoring a large area (effective observation radius: 240 km) very quickly (every 10 minutes). It is one of the most efficient remote sensing devices that can produce large areas of rainfall because it can.

The signal scattered to the target (reflectivity) is related to the size distribution of water droplets within the pulse volume emitted by the weather radar, and since the amount of precipitation falling to the ground is a function of the size distribution of the water droplets, the relation between radar reflectivity and ground precipitation is Using (ZR relation: Z = aR ^b ), we estimate ground precipitation from radar reflectivity.

Recently, due to global warming, heavy rains, typhoons, and hurricanes are increasing in many parts of the world, causing much damage. Here, extreme weather phenomena such as heavy rains, typhoons, lightning strikes and heat waves are called severe weather. In 2007, the Intergovernmental Panel on Climate Change (IPCC) anticipated that global warming will continue, and the frequency of risk meteorological events will increase.

Therefore, there is a need for a technology capable of accurately identifying, detecting, and tracking reflex cells that may cause dangerous weather such as thunderstorms, hail, and heavy rain.

Accordingly, the present invention has been made in an effort to provide a method for detecting a reflective cell associated with a dangerous weather, which can automatically detect a reflective cell that can cause a dangerous weather in a 3D laser reflectivity material and calculate its properties. .

One embodiment of the present invention for achieving the above object is to detect the reflectivity cells associated with the dangerous weather by progressively identifying one-dimensional components, two-dimensional components and three-dimensional components from the grid point components of the 3D CAPPI data Stage 1; And a second step of calculating an attribute of the reflectivity cell associated with the detected dangerous weather.

In this case, the first step may include determining grid points exceeding a predetermined reflectivity threshold in the grid point component of the 3D CAPPI data; Retrieving the x and y coordinates of the determined grid points and determining grid points having the same y coordinate and adjacent x coordinates as one-dimensional components; Searching the y coordinates of the determined one-dimensional components to determine one-dimensional components adjacent to the y-coordinate as two-dimensional components; And searching the z-coordinates of the determined two-dimensional components to determine two-dimensional components adjacent to the z-coordinate as three-dimensional components and detecting a three-dimensional reflectivity cell.

The second step may include filtering the determined three-dimensional reflectivity cell based on a predetermined volume threshold; Projecting the filtered three-dimensional reflectivity cell into a two-dimensional x, y plane to convert it into a two-dimensional reflectivity cell; And calculating attribute information about the filtered 3D reflectance cell and the 2D reflectance cell, respectively.

In addition, the attribute information for the two-dimensional reflectivity cell is the area, the two-dimensional center position, the length of the major axis and minor axis of the representative ellipse, the slope of the major axis, the vertical integral liquid content, the vertical integral liquid content density, the upper vertical integral liquid At least one of a group containing the content, wherein the attribute information for the filtered three-dimensional reflectivity cell is at least one of the group including the three-dimensional center position, volume, maximum reflectance, average reflectance, maximum reflectance and position will be.

The second step may further include displaying a CMAX image and a 2D reflectance cell image using the calculated attribute information.

According to the present invention, it is possible to objectively determine the risk of the observed weather phenomena by automatically identifying the dangerous weather and providing quantitative information. When it is used for launching aircraft or space launch vehicles, it is possible to expect significant budget savings by removing business uncertainty based on the objectively determined risk of meteorological phenomena.

Therefore, the present invention has a very bright market as a basic technology essential for the Meteorological Agency, disaster prevention organization, hydrology related organization, aerospace industry, aircraft industry, and the prospect of commercialization or paid technology transfer is very high due to the high interest of the relevant private organizations. bright.

1 is a flowchart illustrating a method of automatically tracking a reflectivity cell associated with a dangerous weather from three-dimensional radar reflectivity data according to an embodiment of the present invention.
FIG. 2 is a flowchart illustrating a 3D CAPPI generation process shown in FIG. 1.
FIG. 3 is a flowchart illustrating a process of identifying a reflectivity cell associated with the dangerous weather shown in FIG. 1.
4 is a flowchart illustrating a tracking process of the reflectivity cell associated with the dangerous weather shown in FIG. 1.
FIG. 5 is a flowchart illustrating a process of determining a membership function and weights used for tracking a reflectivity cell associated with a dangerous weather.
6 to 10 are exemplary diagrams in which the process of determining the membership function and the weight shown in FIG. 5 is actually applied.
11 and 12 are exemplary views showing the results of actually applying the method of tracking the reflectivity cell associated with the dangerous weather shown in FIG.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings, which will be readily apparent to those skilled in the art to which the present invention pertains. As those skilled in the art would realize, the described embodiments may be modified in various different ways, all without departing from the spirit or scope of the present invention. Further, in order to clearly illustrate the present invention in the drawings, portions not related to the description are omitted. In addition, the same code | symbol is attached | subjected about the similar part throughout the specification. In the following description, well-known functions or constructions are not described in detail since they would obscure the invention in unnecessary detail.

1 is a flowchart illustrating a method of automatically tracking a reflectivity cell associated with a dangerous weather from three-dimensional radar reflectivity data according to an embodiment of the present invention.

As shown in FIG. 1, the method of tracking the reflectivity cell associated with the dangerous weather is divided into three parts, generating a 3D constant altitude plan position indicator (CAPPI) (S10), and identifying the reflectance cell associated with the dangerous weather (S20). And the reflection associated with the dangerous weather is also included in the tracking of the cell (S30).

Here, dangerous weather, which is an extreme weather phenomenon that causes enormous damage such as heavy rains, typhoons, lightning strikes, and heat waves, occurs in a convective storm. Therefore, in this embodiment, the reflectivity cell associated with the dangerous weather will be described as a convective storm.

First, the 3D CAPPI generation step S10 is a step of performing coordinate transformation of radar reflectivity data, and converts radar reflectivity data into a 3D reflectance field of a Cartesian coordinate system.

Then, the step S20 of identifying the reflectance cell associated with the dangerous weather detects the convective storm in the coordinate-converted 3D CAPPI data and provides the quantified property of the detected convective storm.

Subsequently, in the tracking step S30 of the reflectance cell related to the dangerous weather, the track of the convective storm is tracked by analyzing the properties of the detected convective storm using fuzzy logic.

Detailed description of each step will be described with reference to FIGS. 2 to 5.

FIG. 2 is a flowchart illustrating a 3D CAPPI generation process shown in FIG. 1.

As shown in FIG. 2, radar reflectivity data is collected (S101).

Then, the collected radar reflectivity data is displayed as radar volume data on the polar coordinate system (S102).

Subsequently, coordinate transformation is performed on the radar volume data (S103). That is, the radar volume data displayed on the polar coordinate system is converted into the 3D reflection field of the rectangular coordinate system. This process produces three-dimensional CAPPI data.

FIG. 3 is a flowchart illustrating a process of identifying a reflectivity cell associated with the dangerous weather shown in FIG. 1.

As shown in FIG. 3, the convective storm identification process is largely divided into two parts: detecting convective storms in the 3D CAPPI data (S21) and calculating attributes of the detected convective storms (S22). Is done.

This convective storm identification process is based on a threshold method to determine reflectivity cells (convective storms) associated with dangerous weather (thunderstorms, hail, heavy rain, etc.) in three-dimensional radar reflectivity data. Uses the reflectivity and volume thresholds to determine convective storms.

First, the step of detecting convective storms in the 3D CAPPI data (S21) is a process of gradually identifying one-dimensional, two-dimensional and three-dimensional components (storm3D) at grid point elements. (Steps S211 to S214). At this time, the determination of the grid point component uses a preset reflectivity threshold.

Specifically, all grid points having reflectivity exceeding the reflectivity threshold are discriminated from the 3D radar reflectivity data and given with individual identification numbers (identification of elements) (S211).

Then, x and y coordinates of all the grid points are searched to determine grid points having the same y coordinate and adjacent x coordinates as one-dimensional components and assigning identification numbers (S212).

Subsequently, the y-coordinates of all one-dimensional components are searched to determine one-dimensional components adjacent to the y-coordinate as two-dimensional components (S213).

Thereafter, the two-dimensional components adjacent to the z-coordinate are searched in the z-direction (vertical direction) of all the two-dimensional components, and the three-dimensional components storm3D, that is, the three-dimensional reflectance cell, are determined (S214).

Next, calculating the attributes of the determined three-dimensional reflectivity cells (S22) includes filtering the determined three-dimensional reflectivity cells, converting them to two-dimensional reflectivity cells, and calculating respective attribute information from the two-dimensional and three-dimensional reflectance cells. It consists of a process (S221 to S225).

In detail, the 3D reflectivity cell is filtered based on a preset volume threshold (S221), and the 3D component (storm3D) having a volume smaller than the volume threshold is excluded from the final determination (S222), and the volume 3 is larger than the volume threshold. Only the dimensional reflectivity cell is projected onto the xy plane and detected as the 2D reflectance cell storm2D (S223). That is, the determined three-dimensional component storm3D is projected onto two-dimensional x and y planes, and the two-dimensional reflectance is converted into a cell and displayed. At this time, the three-dimensional reflectivity cell having a volume larger than the volume threshold is detected as a convective storm, and the two-dimensional reflectivity cell is represented by an ellipse.

Then, three-dimensional attribute information (three-dimensional center position, volume, maximum reflectance, average reflectance, altitude and position of the maximum reflectance, etc.) and two-dimensional attribute information (area, center position, representative) from three-dimensional and two-dimensional reflectivity cells The long axis and short axis length of the ellipse, the slope of the long axis, the vertical integral liquid content, the vertical integral liquid content density, and the upper vertical integral liquid content are calculated to provide quantitative numerical information (S224).

Next, the detected convective storm is displayed as CMAX (Cloumn Maximum) and storm2D images using the calculated attribute information (S225). Here, the CMAX image expresses only the strongest echo in the column of each volume observation data on the plane.

As such, the convective storm identification process according to the present invention may cause dangerous weather (thunderstorm, hail, heavy rain, etc.) in three-dimensional radar reflectivity data of atmospheric phenomenon (cloud, rain, snow) using hydrologic and weather radar. By automatically detecting the presence of a reflectance cell and deriving its properties, you can objectively determine the risk for observed weather phenomena.

4 is a flowchart illustrating a tracking process of the reflectivity cell associated with the dangerous weather shown in FIG. 1.

As shown in FIG. 4, the tracking process of the reflectivity cell associated with the dangerous weather is divided into two parts, determining a belonging function and a weight of the fuzzy logic (S31), and tracking the reflectivity cell associated with the dangerous weather ( S32).

The tracking process of the reflectivity cell associated with the dangerous weather uses fuzzy logic to track the reflectance cells determined from the radar reflectivity data of different time zones over time, and the fuzzy logic is composed of a membership function and weights.

First, the determining of the membership function and the weight of the fuzzy logic (S31) consists of a process of determining the membership function and the weight.

In detail, the fuzzy parameter of the fuzzy logic is calculated using the attribute information of the 3D reflectance cell Storm3D (S311). Here, the fuzzy parameters of the fuzzy logic include a moving speed (SPD), a difference of mean reflectivity (DMR), an area change ratio (ACR), and an axis transformation ratio (ATR). . The calculation of these fuzzy variables is calculated using the properties of the detected convective storms (mean reflectance, central location, minimum altitude, highest altitude, etc.). Four fuzzy variables are shown in Equations 1 to 4, respectively.

SPD is ㎳ ^-1 unit to calculate a center position of the three-dimensional reflection cell (storm3D) it is determined at two successive times the moving speed is calculated according to equation (1).

,

는 각각 과거시간(t-△t)과 현재시간(t)에 판별한 storm3D의 중심위치의 x좌표를 나타내며,

,

는 각각 과거시간(t-△t)과 현재시간(t)에 판별한 반사도 셀(storm3D)의 중심위치의 y좌표를 나타내며, m, n는 과거시간(t-△t)과 현재시간(t)에 판별한 반사도 셀(storm3D)의 식별번호(ID)를 나타낸다.here,

,

Represents the x-coordinate of the center position of storm3D determined at the past time (t- △ t) and the current time (t), respectively.

,

Denotes the y-coordinate of the center position of the reflectivity cell storm3D determined at the past time t-Δt and the current time t, respectively, where m and n are the past time t-Δt and the current time t Indicates the identification number ID of the reflectance cell storm3D.

The DMR is a change rate of the average reflectivity between the reflectance cells (storm3D) determined at two time zones and a difference in reflectance between the convective cells determined at the present time and the average reflectance of the reflectance cells (storm3D) determined at the current time zone. Defined in dB and calculated according to equation (2).

,

는 각각 과거시간(t-△t)과 현재시간(t)에 판별한 반사도 셀(storm2D)들의 평균 반사도를 나타내며 dBZ 단위이다.here,

,

Denotes the average reflectivity of the reflectivity cells storm2D determined at the past time t-Δt and the current time t, respectively, in dBZ units.

The ACR is defined as a ratio of the area difference between the past and present discriminated reflectivity cells storm2D to the area of the reflectivity cells storm2D determined at the present time and is calculated according to Equation 3 as a dimensionless value. .

,

는 각각 과거시간(t-△t)과 현재시간(t)에 판별한 반사도 셀(storm2D)(storm3D를 2차원으로 투영시킨 것)의 면적으로 km^-2 단위이다. avg는 평균을 의미한다.here,

,

Denotes the area of the reflectivity cell storm2D (projected storm3D in two dimensions) determined at the past time t-Δt and the current time t, respectively, in km- ² units. avg means mean.

In addition, ATR is defined as the rate of change of the long and short reflectivity cells (storm2D) determined in the past and the current time for the long axis and short axis of the ellipse representing the reflectivity cell (storm2D) determined at the present time, and the equation is dimensionless. Calculated according to 4.

,

,

,

는 각각 과거시간(t-△t)과 현재시간(t)에 판별한 반사도 셀(storm2D) 타원의 장축과 단축을 나타내는 km 단위이다. 장축과 단축의 변형비율은 각각 동일한 가중치(1/2)로 가중되어 합산된다. avg는 평균을 의미한다. here,

,

,

,

Denotes a long axis and a short axis of the reflectivity cell (storm2D) ellipse determined at the past time t-Δt and the current time t, respectively. The strain ratios of the long and short axes are each weighted with the same weight (1/2) and summed. avg means mean.

Four fuzzy variables are calculated using the above Equations 1 to 4, and then the membership function and weights are determined using the normalized frequency distribution of the four fuzzy variables calculated (S312).

Next, the total belonging value is calculated by multiplying the belonging value by the weight (S313). This is called fuzzy normalization. Fuzzy normalization consists of membership functions and their weights. Four fuzzy variable values are converted to membership values using the membership function. Then, the total belonging value is calculated as in Equation 5 below.

Here, MV represents a belonging value and W represents a weight.

Then, when the calculated total belonging value exceeds the threshold (MV _tot > MV _thresh ), the dangerous weather determined in two different time zones is determined as the same dangerous weather (S32).

As such, the tracking of the reflectance cell associated with the dangerous weather is implemented by fuzzy logic using four fuzzy variables. Each fuzzy variable is derived from two convective storm characteristics determined at two consecutive times. And, two convective storms at different times are tracked by fuzzy normalization. At this time, the membership function and its weight are objectively determined in advance from the statistical data.

FIG. 5 is a flowchart illustrating a process of determining a membership function and weights used for tracking a reflectivity cell associated with a dangerous weather.

As shown in FIG. 5, radar reflectivity data of different time zones are collected (S50), 3D CAPPI generation (S51) is generated from the collected radar reflectivity data (S51), and the reflectance cells are determined (S52). It is displayed (S53).

A case is selected for the reflectance cell determined as described above (S54), and a fuzzy variable is calculated through statistical processing (S55). In this case, the selected cases are classified into storm lists that do not match and storm lists that do not match.

Then, by using the calculated fuzzy variable objectively calculate the membership function and weight to determine (S56).

6 to 10 are exemplary diagrams in which the process of determining the membership function and the weight shown in FIG. 5 is actually applied.

6 is an exemplary diagram illustrating selecting a convective storm case for determining belonging function and weight.

As shown in FIG. 6, six cases are selected and each convective storm is represented using PNG (Portable Network Graphics) and ASCII.

The convective storm list is matched with the convective storm list that is not matched.

FIG. 7 is a graph showing the normalized frequency distribution of the matched and mismatched convective storm pairs shown in FIG. 6. Here, the black line indicates a matching storm pair and the red line indicates a mismatched storm pair.

As shown in FIG. 7, the frequency of matching storm pairs is concentrated at low values, but the non-matching storm pairs are widely distributed to high values. This means that the characteristics between the two storm pairs are quite different.

On the other hand, two frequency distributions for each fuzzy variable have overlapping regions. The size of this overlap region is different for each fuzzy variable, and the smaller the overlap region of two normalized frequency distributions, the more useful the storm tracking. This is because a large overlap region means that two frequency distributions have similar patterns and characteristics.

Referring to FIG. 7, since the overlap region of the SPD is the smallest, it can be seen that it is an important variable for storm tracking than other fuzzy variables.

8 is an exemplary diagram illustrating a process of determining a membership function for the ATR of FIG. 7.

Here, the membership function is defined as the ratio of one frequency to the sum of two frequencies, and is obtained according to Equation 6 below.

Where MF is the membership function and F is the frequency for the fuzzy variable.

When the membership function is calculated according to Equation 6 from the normalized frequency distribution of the ATR shown in FIG.

FIG. 9 is an exemplary diagram illustrating a membership function determined from normalized frequency distributions of four fuzzy variables shown in FIG. 7. Here, the black line indicates the belonging function of the matching storm pair, and the red line indicates the belonging function of the unmatched storm pair.

As shown in FIG. 9, all membership functions of the matching storm pair decrease as the fuzzy variable value increases. If a high value is calculated between two storms, the two storms do not match as the same storm.

FIG. 10 is an exemplary diagram illustrating a process of determining weights for belonging functions of four fuzzy variables of FIG. 7.

Here, the weight is related to the overlapping region of two normalized frequency distributions between the matching and unmatched storm pairs, and is inversely related to the size of the overlapping region, and is calculated according to Equation 7 below.

As shown in FIG. 10, since the SPD has the smallest overlap region, the SPD has a weight of 0.6 and has the highest value among four fuzzy variables, and the DMR has the smallest weight of 0.11.

This result indicates that SPD is the most important variable among four fuzzy variables in storm tracking.

11 and 12 are exemplary views showing a result of actually applying the storm tracking method according to the present invention.

FIG. 11 is an exemplary diagram illustrating a CMAX image and a storm2D image and a storm tracking result for the time of July 18, 2008 at about 12:05.

As shown in (a) of FIG. 11, according to the CMAX image on the left side and the storm2D image on the right side, it can be seen that several storms are generated around the radar and are moving in the northeast direction.

As shown in (b) of FIG. 11, reference storm tracks G1 to G4 are listed in the left table, and according to elliptical images of reference storm tracks superimposed for seven times, the same color is changed over time. According to the same storm. The red line indicates storm tracking by the storm tracking method according to the present invention.

Storm movements occur simultaneously with each other. Although G2 and G3 occur closely, the storm tracking method according to the present invention can accurately track.

FIG. 12 is an exemplary diagram illustrating a CMAX image and a storm2D image and a storm tracking result for the same time on July 18, 2008, at 05:35.

As shown in (a) of FIG. 12, according to the CMAX image on the left side and the storm2D image on the right side, it can be seen that some small storms are generated around the radar and are moving in the northwest direction.

As shown in FIG. 12B, reference storm tracks G1 to G8 are listed in the left table, and according to an ellipse image of the reference storm track superimposed for 10 times, the movement of the storm (track) Occur simultaneously with each other. That is, even if a large number of storms are generated, the storm tracking method according to the present invention can accurately track the course.

So far I looked at the center of the preferred embodiment for the present invention. 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.

In addition, 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 can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion.

Claims (7)

A first step of gradually detecting one-dimensional components, two-dimensional components, and three-dimensional components from the lattice point components of the 3D CAPPI data to detect reflectance cells associated with dangerous weather; And
And calculating a property of the reflectance cell associated with the detected dangerous weather. 2.
The method of claim 1, wherein the first step,
Determining grid points that exceed a predetermined reflectivity threshold in the grid point component of the 3D CAPPI data;
Retrieving the x and y coordinates of the determined grid points and determining grid points having the same y coordinate and adjacent x coordinates as one-dimensional components;
Searching the y coordinates of the determined one-dimensional components to determine one-dimensional components adjacent to the y-coordinate as two-dimensional components; And
Detecting the 3D reflectivity cell by searching the z coordinates of the determined 2D components and determining adjacent 2D components as 3D components. .
The method of claim 2, wherein the second step,
Filtering the determined three-dimensional reflectivity cell based on a preset volume threshold;
Projecting the filtered three-dimensional reflectivity cell into a two-dimensional x, y plane to convert it into a two-dimensional reflectivity cell; And
And calculating the filtered three-dimensional reflectivity cell and attribute information for the two-dimensional reflectivity cell, respectively.
The method of claim 3,
The attribute information for the two-dimensional reflectivity cell includes the area, the two-dimensional center position, the length of the major axis and the minor axis of the representative ellipse, the slope of the major axis, the vertical integral liquid content density, the vertical integral liquid content density, and the upper vertical integral liquid content. At least one of a group comprising a reflectance cell associated with the dangerous weather.
The method of claim 3,
The attribute information of the filtered three-dimensional reflectivity cell is at least one of a group including a three-dimensional center position, volume, maximum reflectance, average reflectivity, the height and position of the maximum reflectivity of the reflectivity cell associated with the dangerous weather Detection method.
The method of claim 3,
The second step may further include displaying a CMAX image and a 2D reflectance cell image using the calculated attribute information.
A recording medium having recorded thereon a program which can be read and executed by a digital signal processing apparatus as a program for performing the method of any one of claims 1 to 6.

Images