Information processing device, weather radar system, analysis and synthesis processing station, method and program

Abstract

The objective is to provide an information processing device, a weather radar system, an analysis and synthesis processing station, a method, and a program that can detect the occurrence of characteristic precipitation areas in advance. [Solution] The information processing device according to the embodiment includes a processing unit that detects precipitation areas based on the first hourly cumulative value of rainfall at each of the multiple locations, which is based on upper-air rainfall data indicating hourly rainfall above the multiple locations, and detects the occurrence of characteristic precipitation areas based on statistical values ​​of rainfall calculated for the detected precipitation areas.

Description

[Technical Field]

[0001] Embodiments of the present invention relate to an information processing device, a weather radar system, an analysis and synthesis processing station, a method, and a program. [Background technology]

[0002] In recent years, it has become known that severe damage can occur in characteristic precipitation areas, such as linear rainbands.

[0003] To mitigate such damage, it is useful to detect the occurrence of the characteristic precipitation areas described above in advance. [Prior art documents] [Patent Documents]

[0004] [Patent Document 1] Patent No. 6689396 [Overview of the project] [Problems that the invention aims to solve]

[0005] Therefore, the problem that the present invention aims to solve is to provide an information processing device, a weather radar system, an analysis and synthesis processing station, a method, and a program that can detect the occurrence of characteristic precipitation areas in advance. [Means for solving the problem]

[0006] The information processing device according to the embodiment includes a processing unit that detects precipitation areas based on the first hourly cumulative value of rainfall at each of the multiple locations, which is based on upper-air rainfall data indicating hourly rainfall above the multiple locations, and detects the occurrence of characteristic precipitation areas based on statistical values ​​of rainfall calculated for the detected precipitation areas. [Brief explanation of the drawing]

[0007] [Figure 1]Block diagram showing an example of the functional configuration of the information processing apparatus according to the embodiment. [Figure 2] Diagram showing an example of the system configuration of the information processing apparatus. [Figure 3] Diagram for explaining an example of upper air rainfall data. [Figure 4] Flowchart showing an example of the processing procedure of the information processing apparatus. [Figure 5] Diagram for explaining the precipitation area. [Figure 6] Diagram for explaining the rainfall statistical value. [Figure 7] Diagram showing an example of the display of the detection result. [Figure 8] Block diagram showing an example of the information processing apparatus according to the modification of the present embodiment. [Figure 9] Diagram for explaining an example of the upwind area. [Figure 10] Diagram showing another example of the display of the detection result. [Figure 11] Block diagram showing another example of the functional configuration of the information processing apparatus according to the modification of the present embodiment. [Figure 12] Diagram for explaining the feature amount calculated from the prediction data. [Figure 13] Diagram showing an example of the feature amount input to the first machine learning model. [Figure 14] Diagram showing an example of the feature amount input to the second machine learning model. [Figure 15] Diagram showing an example of the configuration of the weather radar system. [Figure 16] Diagram showing an example of the configuration of the radar analysis / synthesis processing unit.

Mode for Carrying Out the Invention

[0008] Hereinafter, the embodiment will be described with reference to the drawings. FIG. 1 is a block diagram showing an example of the functional configuration of the information processing apparatus according to the present embodiment. As shown in FIG. 1, the information processing apparatus 10 includes a rainfall data storage unit 11 and a processing unit 12.

[0009] The rainfall data storage unit 11 stores rainfall data indicating the rainfall for each time at a plurality of locations. The rainfall data stored in the rainfall data storage unit 11 is data indicating rainfall generated based on data observed in a weather radar system (hereinafter referred to as observation data) not shown in the figure, for example, and is acquired from the weather radar system. Also, in the weather radar system, as a weather radar for acquiring observation data, for example, a parabolic weather radar, a multi-parameter phased array weather radar (MP-PAWR), a phased array weather radar (PAWR), or the like may be adopted, or other types of weather radars may be adopted.

[0010] The processing unit 12 includes a precipitation area detection unit 121, a statistical value calculation unit 122, and a detection unit 123, and executes processing for detecting in advance the occurrence of a characteristic precipitation area using the rainfall data stored in the rainfall data storage unit 11. In other words, it can be said that the information processing apparatus 10 according to the present embodiment has a function as a weather prediction apparatus for predicting the occurrence of a characteristic precipitation area.

[0011] Note that the characteristic precipitation area in the present embodiment corresponds to a rain area (dangerous precipitation area) accompanied by strong precipitation where damage occurs, and is, for example, an area (region) affected by a linear precipitation band, local heavy rain, or concentrated heavy rain.

[0012] Hereinafter, it will be described as detecting the occurrence of a linear precipitation band. Note that a linear precipitation band can be defined as a rain area with strong precipitation that extends linearly, about 50 to 300 km in length and about 20 to 50 km in width, created by an organized cumulonimbus cloud group in which developed rain clouds (cumulonimbus clouds) that occur one after another form a line and pass through or stagnate at substantially the same location for several hours.

[0013] Here, it is assumed that the rainfall data stored in the rainfall data storage unit 11 in the present embodiment is three-dimensional rainfall data indicating the rainfall at each grid point in the real space (three-dimensional space) defined by latitude, longitude, and altitude.

[0014] In this case, the precipitation area detection unit 121 can generate upper-air rainfall data showing hourly rainfall above multiple locations based on the three-dimensional rainfall data described above. In this embodiment, the upper-air rainfall data corresponds, for example, to data showing rainfall in the range from the ground surface to the upper atmosphere at multiple locations.

[0015] The precipitation area detection unit 121 obtains a value obtained by accumulating the rainfall at each of multiple locations over time, based on the generated upper-air rainfall data, and detects an area containing multiple locations with large values ​​as a precipitation area.

[0016] The statistical value calculation unit 122 calculates statistical values ​​of rainfall (hereinafter referred to as rainfall statistics) for each of the multiple points included in the precipitation area detected by the precipitation area detection unit 121.

[0017] The detection unit 123 detects that a linear rainband will occur in the future (hereinafter referred to as "occurrence of a linear rainband") based on the rainfall statistics calculated by the statistical value calculation unit 122. In this embodiment, detecting the occurrence of a linear rainband corresponds to determining (distinguishing) whether the precipitation area detected by the precipitation area detection unit 121 described above is a precipitation area that leads to a linear rainband.

[0018] Figure 2 shows an example of the system configuration of the information processing device 10 shown in Figure 1. The information processing device 10 includes a CPU 10a, non-volatile memory 10b, RAM 10c, and a communication device 10d, etc.

[0019] The CPU 10a is a processor for controlling the operation of various components within the information processing device 10. The CPU 10a may be a single processor or may consist of multiple processors. The CPU 10a executes various programs loaded from the non-volatile memory 10b into the RAM 10c. The programs executed by the CPU 10a include weather forecasting programs for detecting the occurrence of linear precipitation bands (i.e., predicting the weather).

[0020] The non-volatile memory 10b is a storage medium used as an auxiliary storage device. The RAM 10c is a storage medium used as a main storage device. Although only the non-volatile memory 10b and RAM 10c are shown in Figure 2, the information processing device 10 may also include other storage devices such as an HDD (Hard Disk Drive) and an SSD (Solid State Drive).

[0021] The communication device 10d is a device configured to perform wired or wireless communication.

[0022] Although not shown in Figure 2, the information processing device 10 may include, for example, input devices such as a mouse and keyboard, and display devices such as a display.

[0023] In this embodiment, the rainfall data storage unit 11 shown in Figure 1 is implemented by, for example, a non-volatile memory 10b or another storage device.

[0024] Furthermore, in this embodiment, the processing unit 12 shown in Figure 1 is implemented by at least one processor. The processor includes, for example, a control unit and an arithmetic unit, and is implemented by analog or digital circuits. The processor may be the CPU 10a described above, or it may be a general-purpose processor, a microprocessor, a digital signal processor (DSP), an ASIC, an FPGA, or a combination thereof.

[0025] Furthermore, part or all of the processing unit 12 may be implemented by having the CPU 10a (i.e., the computer of the information processing device 10) execute the weather forecast program described above, that is, by software. This weather forecast program may be stored and distributed on a computer-readable storage medium, or it may be downloaded to the information processing device 10 via a network. Part or all of the processing unit 12 may also be implemented by dedicated hardware, etc.

[0026] Here, with reference to Figure 3, an example of the above-mentioned upper-air rainfall data will be explained. As shown in Figure 3, the upper-air rainfall data in this embodiment is data in which the amount of rainfall at each of the multiple points placed on the map is assigned to that point.

[0027] Furthermore, the rainfall amounts assigned to each of the multiple points in the upper-air rainfall data (i.e., the rainfall amounts indicated by the upper-air rainfall data) correspond to the rainfall amounts in the range from the ground surface to the upper atmosphere, as described above, and are, for example, the cumulative value of the rainfall amounts at each grid point defined within that range (i.e., in the height direction). Specifically, the upper-air rainfall data in this embodiment is data that represents the vertically integrated liquid water content (VIL), and this VIL can be calculated based on the three-dimensional rainfall data described above.

[0028] In such upper-air rainfall data, for example, each of multiple locations can be represented by a color corresponding to the amount of rainfall. Each of the multiple locations placed on the map in the upper-air rainfall data corresponds to a grid point represented by the latitude and longitude defined in that map.

[0029] Figure 3 also shows, for example, upper-air rainfall data indicating the amount of rainfall over a predetermined period of time. The rainfall data storage unit 11 stores, for example, three-dimensional rainfall data acquired from the weather radar system from the past to the present, at intervals such as 5 minutes. In this embodiment, upper-air rainfall data for each hour is prepared based on this three-dimensional rainfall data.

[0030] Below, an example of the processing procedure of the information processing device 10 according to this embodiment will be described with reference to the flowchart in Figure 4.

[0031] First, the precipitation area detection unit 121 included in the processing unit 12 acquires three-dimensional rainfall data stored in the rainfall data storage unit 11 (step S1).

[0032] Next, the precipitation area detection unit 121 calculates the VIL (vertically integrated rainwater volume) at multiple locations based on the three-dimensional rainfall data acquired in step S1 (step S2).

[0033] The rainfall data storage unit 11 stores hourly 3D rainfall data, and the processes in steps S1 and S2 described above are performed for each of the hourly 3D rainfall data.

[0034] In other words, when steps S1 and S2 are executed, the precipitation area detection unit 121 acquires hourly VIL data that shows the VIL calculated at multiple locations based on the three-dimensional rainfall data. This VIL data is an example of the upper-air rainfall data described above.

[0035] Next, the precipitation area detection unit 121 calculates the time-integrated VIL value (hereinafter referred to as time-integrated VIL) at one of the multiple locations (hereinafter referred to as the target location) based on the VIL data for each hour that corresponds to a predetermined time period (for example, the time period from the current time to 60 minutes prior) from the hourly VIL data described above (step S3). The time-integrated VIL calculated in step S3 is a value obtained by sequentially adding up the rainfall at the target location extracted from the hourly VIL data (i.e., integrating over time). Specifically, time-integrated VIL ts (kg / m 2 ) is calculated by the following formula (1).

number

[0036] In equation (1) above, T is the cumulative time, VIL t Δt represents the instantaneous value of VIL at time t, and Δt represents the observation frequency of the instantaneous value of VIL accumulated over time. In this case, the accumulation time T is assumed to be, for example, 60 minutes or 180 minutes, and the observation frequency Δt is assumed to be, for example, 1 minute.

[0037] When the process in step S3 is executed, it is determined whether the process in step S3 has been executed for all of the above-mentioned locations (i.e., whether the time-integrated VIL has been calculated for all locations) (step S4).

[0038] If it is determined that the time-integrated VIL has not been calculated at any point (NO in step S4), the process returns to step S3 and is repeated. In this case, the process in step S3 is executed with the points for which the time-integrated VIL has not been calculated as target points.

[0039] On the other hand, if it is determined that the time-integrated VIL has been calculated at all locations (YES in step S4), the precipitation area detection unit 121 detects a precipitation area that includes at least some of the locations based on the time-integrated VIL at each of the locations calculated by repeatedly executing the process in step S3 (step S5). In step S5, multiple precipitation areas may be detected.

[0040] Now, with reference to Figure 5, the precipitation area detected in step S5 will be described. Figure 5 shows an example of a 1-hour time-integrated VIL at each of several locations (i.e., grid points represented by latitude and longitude).

[0041] In this embodiment, for example, a region (a group of grid points) containing points where the time-integrated VIL is greater than or equal to a threshold is detected as a precipitation area. The threshold for the time-integrated VIL is, for example, 30 kg / m 2 According to this, the time-integrated VIL shown in Figure 5 is 30 kg / m 2 The region 121a, which includes the above-mentioned location, can be detected as a precipitation area.

[0042] Here, we have described that an area containing points where the time-integrated VIL is above a threshold is detected as a precipitation area, but it is preferable to also set a threshold for the area of ​​the said area. This area threshold could be, for example, 50 km. 2 According to this, for example, region 121b shown in Figure 5 (that is, area of ​​50 km²)2 Regions less than that are not detected as precipitation regions, and very small regions can be excluded from the precipitation regions.

[0043] For example, in the Japan Meteorological Agency, an area where the 3-hour integrated rainfall value is 100 mm (33 mm / h) or more and the area is 500 km 2 or more is detected as a linear precipitation band. However, in the present embodiment, in order to detect the linear precipitation band earlier, a configuration is adopted in which the precipitation region is detected with a shorter integration time and a smaller area.

[0044] Also, here, it has been described that a region including a point where the time-integrated VIL is 30 kg / m 2 or more is detected as a precipitation region. However, for example, the time-integrated VIL at all points included in the detected precipitation region does not have to be 30 kg / m 2 or more. Specifically, as shown in FIG. 5, a region 121c including a point where the time-integrated VIL is equal to or greater than a first threshold value (for example, 30 kg / m 2 ) and a point adjacent to the point where the time-integrated VIL is equal to or greater than a second threshold value (for example, 20 kg / m 2 ) may be detected as a precipitation region. In this case, a region that does not have a point where the time-integrated VIL is equal to or greater than the first threshold value and includes only points where the time-integrated VIL is equal to or greater than the second threshold value is not detected as a precipitation region.

[0045] Furthermore, in the example shown in FIG. 5, regions 121d and 121e are detected as precipitation regions in the same manner as the above-described region 121c. However, when a plurality of regions are located in the vicinity (the distance between the plurality of regions is equal to or less than a threshold value), such as regions 121d and 121e, a region 121f including the regions 121d and 121e may be detected as one precipitation region. In this case, regions 121d, region 121e, and region 121f may be detected as different precipitation regions, respectively.

[0046] Furthermore, as described above, the detected precipitation area may be corrected to approximate a predetermined shape. Note that the region 121f shown above illustrates an example where the region including regions 121d and 121e is corrected to an ellipse shape. The corrected shape of the precipitation area may be any shape other than an ellipse.

[0047] Returning to Figure 4, the statistical value calculation unit 122 calculates statistical values ​​for the precipitation area detected in step S5 (hereinafter referred to as the target precipitation area) (step S6). In step S6, for example, statistical values ​​for rainfall at each of the multiple locations included in the target precipitation area (hereinafter referred to as rainfall statistics) are calculated.

[0048] The following explains the rainfall statistics calculated as statistical values ​​for the target precipitation area, referring to Figure 6.

[0049] First, the statistical value calculation unit 122 calculates the time-integrated VIL at a given location by accumulating the VIL at each of the multiple locations indicated by the VIL data for each time.

[0050] Next, the statistical value calculation unit 122 extracts (acquires) the time-integrated VIL for each point included in the target precipitation area from among the multiple points for which the time-integrated VIL has been calculated as described above.

[0051] Furthermore, the statistical value calculation unit 122 calculates rainfall statistics based on the extracted time-integrated VIL. In this case, the statistical value calculation unit 122 calculates the maximum value, average value, or total value of the time-integrated VIL in the target precipitation area (a two-dimensional space corresponding to an area defined by latitude and longitude).

[0052] Generally, the average value of the time-integrated VIL in a target precipitation area is calculated, for example, by dividing the sum of the time-integrated VILs at each point included in the target precipitation area by the number of such points. However, the number of points to which this sum is divided may be the number of points among the multiple points included in the target precipitation area where the time-integrated VIL is not 0 (i.e., grid points where there are rainfall amounts greater than 0).

[0053] Returning to Figure 4, the detection unit 123 performs a process to detect the occurrence of a linear rainband based on the rainfall statistics calculated in step S6. In this case, the detection unit 123 determines whether the target precipitation area described above is a rainband connected to a linear rainband (step S7).

[0054] Specifically, in step S7, a process is performed to compare the rainfall statistics calculated in step S6 with a pre-prepared threshold. For example, if the rainfall statistics are equal to or greater than the threshold, it is determined that the target precipitation band is a precipitation band connected to a linear precipitation band.

[0055] Here, we assume that the target precipitation area is determined to be a precipitation area that will lead to a linear precipitation band (YES in step S7). The fact that the target precipitation area is a precipitation area that will lead to a linear precipitation band means that a linear precipitation band will occur in the future, and the detection unit 123 can detect the occurrence of the linear precipitation band.

[0056] If the occurrence of a linear rainband is detected in this manner, the detection result is output (step S8). In step S8, for example, the detection result may be output to the communication device 10d in order to transmit the detection result to an external device, or the detection result may be output to the display device in order to display the detection result on the display device. The detection result output in step S8 only needs to be information indicating that a linear rainband will occur in the future, but for example, as shown in Figure 7, by outputting information that represents the target precipitation area 123a, which is a precipitation area connected to a linear rainband, on a map as the detection result, users using this information can intuitively grasp the area where the occurrence of a linear rainband is predicted. In addition, the detection result may be processed into information to be distributed to the user before the processing in step S8 is executed.

[0057] On the other hand, if it is determined that the target precipitation area is not a precipitation area that leads to a linear precipitation band (NO in step S7), the process in step S8 is omitted, and the process shown in Figure 4 ends. Here, the explanation assumes that the process in step S8 is omitted, but if it is determined that the target precipitation area is not a precipitation area that leads to a linear precipitation band, information indicating that a linear precipitation band will not occur (or is unlikely to occur) may be output.

[0058] Although not shown in Figure 4, if multiple precipitation areas are detected in step S5 as described above, the processes in steps S6 to S8 will be repeatedly executed with each of the multiple precipitation areas as the target precipitation area.

[0059] As described above, the information processing device 10 (processing unit 12) according to this embodiment detects precipitation areas based on the time-integrated VIL (first hourly integrated value of rainfall) at each of the multiple locations, which is based on VIL data indicating the hourly VIL (vertical integrated rainfall) at the multiple locations, and detects the occurrence of linear rainbands (characteristic precipitation areas) based on statistical values ​​(e.g., rainfall statistics) calculated with respect to the detected precipitation areas.

[0060] Generally, the Japan Meteorological Agency (JMA) announces the occurrence of a linear rainband, for example, as part of its information on such bands. However, such announcements by the JMA are based on, for example, three-hour cumulative rainfall data and are equivalent to real-time information. In other words, when the JMA announces the occurrence of a linear rainband, heavy rainfall has often continued for a certain period of time, and damage (disaster) may have already occurred.

[0061] In contrast, in this embodiment, with the above-described configuration, if, for example, there is 3D rainfall data (VIL data indicating VIL calculated from it) for about 60 minutes, the occurrence of a linear rainband can be detected in advance (for example, several hours before the Japan Meteorological Agency makes an announcement).

[0062] Furthermore, the VIL data described in this embodiment is an example of upper-air rainfall data showing hourly rainfall above multiple locations. In this embodiment, by using such VIL data, the system considers not only the amount of rainfall at the ground (ground rainfall) but also information on rain that may fall on the ground in the future. With this configuration, it is expected that signs of linear rainbands can be detected earlier compared to using only ground rainfall.

[0063] Furthermore, predicting the location and amount of rainfall of linear rainbands is difficult, making it impossible to accurately estimate the amount of rainfall flowing into rivers. This has impacted river management, for example, using water level prediction models. However, in this embodiment, as described above, it is possible to detect the occurrence of linear rainbands by capturing early signs of them, and to provide the detection results to users as useful information for river management. Specifically, as described above, if the occurrence of a linear rainband is detected by determining that the target rainband is connected to a linear rainband, it is possible to issue warnings regarding rising water levels in rivers within the basin overlapping with the target rainband. The detection results may also be used as useful information for disaster prevention by issuing warnings to residents of municipalities overlapping with the target rainband.

[0064] Furthermore, in this embodiment, by using a phased array weather radar (MP-PAWR or PAWR) to generate VIL data (upper-air rainfall data), it is possible to further improve the detection accuracy of linear precipitation bands. Although this embodiment assumes the use of a phased array weather radar, a parabolic weather radar or other weather radar may also be used.

[0065] In this embodiment, a precipitation area is detected based on the time-integrated VIL at each of several locations, and rainfall statistics are calculated for each of the time-integrated VILs at several locations included in the precipitation area. However, the integration time used to calculate the rainfall statistics may differ from the integration time used to detect the precipitation area. With such a configuration, for example, a relatively long integration time can be used to detect areas where rainfall has occurred for a long period of time as a precipitation area, and a relatively short integration time can be used to detect the occurrence of a linear rainband when it is determined that there has been heavy rainfall in a short period of time. This makes it possible to detect the occurrence of a linear rainband early with high accuracy.

[0066] Furthermore, in this embodiment, the rainfall statistics used to detect the occurrence of linear rainbands were described as being calculated based on the time-integrated VIL in the target precipitation area. However, a configuration in which multiple integration times are prepared for calculating the said time-integrated VIL is also possible.

[0067] Specifically, assuming that a first and second cumulative time are provided as the multiple cumulative time options mentioned above, the system can be configured to detect the occurrence of a linear rainband based on a first rainfall statistic calculated from the time-cumulative VIL in the target precipitation area based on the first cumulative time (e.g., 60 minutes) and a second rainfall statistic calculated from the time-cumulative VIL in the target precipitation area based on the second cumulative time (e.g., 180 minutes).

[0068] With this configuration, linear precipitation areas (rainfall areas) are less likely to form over short accumulation periods, while over long accumulation periods, precipitation areas that lead to linear precipitation bands may appear even if there has been little rainfall recently. Therefore, by combining rainfall statistics based on multiple accumulation periods, the accuracy of detecting the occurrence of linear precipitation bands can be improved.

[0069] By the way, in this embodiment, a configuration for detecting the occurrence of linear rainbands using VIL data (3D rainfall data) has been described, but it is also possible to use a configuration for detecting the occurrence of linear rainbands using data other than VIL data (hereinafter referred to as a modified version of this embodiment).

[0070] Modifications of this embodiment will be described below. Figure 8 is a block diagram showing an example of the functional configuration of the information processing device 10 according to a modification of this embodiment. In Figure 8, parts that are the same as those in Figure 1 described above are denoted by the same reference numerals and their detailed descriptions are omitted, while parts that differ from Figure 1 will be described.

[0071] As shown in Figure 8, the information processing device 10 includes a wind condition data storage unit 13. The wind condition data storage unit 13 stores wind condition data, for example, showing the wind speed and wind direction for each hour at multiple locations. The wind condition data stored in the wind condition data storage unit 13 may be acquired from a system other than the weather radar system (for example, a weather observation system).

[0072] In a modified version of this embodiment, when detecting the occurrence of a linear precipitation band, statistical values ​​of wind speed and wind direction (hereinafter referred to as wind condition statistics) calculated from the wind condition data stored in the wind condition data storage unit 13 may be used. The wind condition statistics are assumed to be, for example, the average wind speed or average wind direction in the target precipitation area, but they may also be the angle (angle difference) between the average wind direction and the direction of travel (movement) of the target precipitation area.

[0073] Furthermore, if the wind condition data is, for example, data indicating wind speed and wind direction in a three-dimensional space (i.e., three-dimensional grid data of latitude, longitude, and altitude), then the wind condition statistics may be calculated for each of the multiple altitude planes, or they may be the difference or dot product of the average wind direction between the multiple altitude planes in the target precipitation area.

[0074] The wind condition statistics may be one of the values ​​described here, or a combination of two or more values.

[0075] In a modified version of this embodiment, for example, in addition to the rainfall statistics described in this embodiment, the wind condition statistics described above may be used to detect the occurrence of linear rainbands. In this case, for example, if it is determined that winds strong enough to develop into a precipitation area connected to a linear rainband are occurring within the target precipitation area, the occurrence of a linear rainband can be detected. With such a configuration, it is possible to improve the accuracy of detecting the occurrence of linear rainbands by considering, for example, wind information near the ground and in the upper atmosphere.

[0076] Furthermore, it is generally known that rain clouds (cumulonimbus clouds) move as they are carried by the wind (that is, they move from upwind to downwind), and in order to detect the occurrence of linear rainbands, for example (that is, to determine whether or not the target precipitation area is a precipitation area connected to a linear rainband), it is considered useful to consider information about the area located upwind of the target precipitation area (hereinafter referred to as the upwind area).

[0077] In this case, in addition to the statistical values ​​for the target precipitation area mentioned above (rainfall statistics or wind condition statistics for the target precipitation area), the accuracy of detecting the occurrence of linear precipitation bands can be improved by further utilizing statistical values ​​for the upwind area identified based on wind condition data, for example (rainfall statistics or wind condition statistics for the said upwind area).

[0078] An example of the upwind region will be described with reference to Figure 9. In this modified embodiment, the upwind region 202 of the target precipitation area 201 is defined as, for example, a region within a distance of 150 km from the center of the target precipitation area 201, and within an angle of 30° centered on the upwind vector.

[0079] Here, we have described a case where the upwind region is statically determined according to the location of the target precipitation area. However, the upwind region may also be dynamically determined based on wind condition data, for example. In this case, the upwind region may be adaptively determined by, for example, determining the area of ​​rain clouds that are expected to reach the location of the target precipitation area within a specified time, based on the average wind speed in the upwind region of the target precipitation area. Specifically, for example, if the average wind speed is 20 m / s and the specified time is 1 hour, the upwind region can be determined to be the area within 72 km upwind from the center of the target precipitation area.

[0080] In a modified version of this embodiment, as described above, it is thought that the lead time can be extended by utilizing statistical values ​​related to the upwind region (focusing on information upwind of the rain area). The lead time is an evaluation value defined by "the time when the linear rainband occurred (for example, the time when the occurrence of the linear rainband was announced by the Japan Meteorological Agency) - the time when the occurrence of the linear rainband was detected," and the longer the lead time, the more it can be evaluated that the occurrence of the linear rainband was detected (predicted) well in advance of the actual time when it occurred.

[0081] Although this explanation assumes the use of wind data in the process of detecting the occurrence of linear precipitation bands, this wind data (wind statistics) may also be used to detect the precipitation areas described above. In this case, for example, areas where wind speed or wind direction meet certain conditions may be detected as precipitation areas.

[0082] Furthermore, for example, when outputting information in which the target precipitation area 123a is represented on a map, as explained in Figure 7, a portion of the target precipitation area 123a, specifically area 123b, may be further represented on the map as a high-risk area, as shown in Figure 10. This area 123b is, for example, an area predicted to be affected by heavy rainfall at the time when a linear precipitation band is expected to occur, due to the movement of rain clouds in area 123c within the target precipitation area 123a, where the time-integrated VIL is large at the time the occurrence of a linear precipitation band is detected, according to the wind speed and wind direction within the target precipitation area 123a indicated by the wind condition data.

[0083] In other words, if the system is configured to utilize wind condition data as described above, it may output area information indicating a portion of the target precipitation area (for example, a dangerous area) based on VIL data indicating the VIL at each of the multiple points included in the target precipitation area, and wind condition data indicating wind speed and wind direction.

[0084] With this configuration, it becomes possible to use information on early detection results of linear rainbands and wind information to narrow down and provide users with information on areas that may become dangerous in the future.

[0085] Figure 11 is a block diagram showing another example of the functional configuration of the information processing device 10 according to a modified example of this embodiment. In Figure 11, parts that are the same as those in Figure 1 are given the same reference numerals and their detailed descriptions are omitted, while parts that differ from Figure 1 will be described.

[0086] As shown in Figure 11, the information processing device 10 includes a prediction data storage unit 14. The prediction data storage unit 14 stores prediction data that shows, for example, the predicted hourly rainfall at multiple locations (hereinafter referred to as predicted rainfall). The prediction data stored in the prediction data storage unit 14 is assumed to be data generated by an external system, for example. Furthermore, the predicted rainfall shown by the prediction data may be VIL (vertical integrated rainfall) or ground rainfall.

[0087] In a modified version of this embodiment, the rainfall statistics calculated from the prediction data stored in the prediction data storage unit 14 can be used as statistics for the target precipitation area to detect the occurrence of a linear precipitation band.

[0088] With this configuration, by considering the predicted rainfall indicated by the forecast data, it is possible to improve the accuracy of detecting the occurrence of linear rainbands and to extend the lead time mentioned above.

[0089] In the modified version of this embodiment, a configuration using wind condition data and forecast data to detect the occurrence of linear rainbands has been described. However, other data may be used if it is useful in improving the accuracy of detecting the occurrence of such linear rainbands. Specifically, in the modified version of this embodiment, for example, information such as polarization information from weather radar, radiosondes, microwave radiometers, aircraft / ship / satellite observation data, and numerical prediction models may be used. In this embodiment, for example, it is determined whether the target rainband is a precipitation area connected to a linear rainband by comparing rainfall statistics with a threshold value. However, a process may be performed to change, modify, or correct the threshold used in the determination process based on, for example, polarization information from weather radar.

[0090] By the way, in the above-described embodiment and its modified form, the occurrence of linear rainbands has been described as being detected based on statistical values ​​related to the target rainband, but a configuration using a machine learning model to detect the occurrence of said linear rainbands is also acceptable.

[0091] In this case, the detection unit 123 maintains a pre-prepared machine learning model and detects the occurrence of a linear rainband by inputting features related to the target precipitation area into the machine learning model (determining whether the target precipitation area is a precipitation area connected to a linear rainband). For example, rainfall statistics for the target precipitation area can be used as features related to the target precipitation area.

[0092] The following describes the machine learning models used to detect the occurrence of linear rainbands. For example, the occurrence of linear rainbands is announced by the Japan Meteorological Agency, but the precipitation area connected to the linear rainband mentioned above is defined as the precipitation area that exists near the location where the linear rainband is to occur before the occurrence of the linear rainband (before the announcement by the Japan Meteorological Agency).

[0093] In this case, the machine learning model is assumed to be pre-generated (prepared) by using past VIL data (3D rainfall data) to learn that the precipitation areas near the linear rainband are connected to the linear rainband, for example, by tracing back 5 minutes at a time starting from the time when the linear rainband occurred in the past.

[0094] Specifically, for example, if the VIL data for the time when the Japan Meteorological Agency announces the occurrence of a linear rainband is taken as the VIL data for time t=0, then a precipitation area that overlaps with the linear rainband, at least partially, is detected based on the VIL data for time t=-5 (i.e., 5 minutes before the occurrence of the linear rainband). The process for detecting this precipitation area is the same as the process in step S5 described above, and the detected precipitation area includes multiple locations where the time-integrated VIL is equal to or greater than a threshold. Furthermore, a precipitation area that overlaps with the linear rainband, at least partially, may be, for example, a precipitation area that overlaps with the linear rainband by a predetermined percentage or more, or a precipitation area whose distance from the linear rainband (or its center or end) is less than or equal to a predetermined value.

[0095] As described above, if the detected precipitation area is used as the training precipitation area, features related to the training precipitation area (e.g., rainfall statistics for the training precipitation area) are calculated, and the machine learning model learns the combination of these features and labels, using these features as explanatory variables and a label indicating that the training precipitation area is connected to a linear precipitation band (i.e., a linear precipitation band occurred) (e.g., label "1") as the training label. The rainfall statistics for the training precipitation area are the maximum, average, or total value of the time-cumulative rainfall in the training precipitation area, and correspond to the rainfall statistics calculated as features related to the target precipitation area as described above.

[0096] Here, we have described the case of training a machine learning model using VIL data at time t=-5. However, this training can be further performed using VIL data from earlier times (for example, VIL data at time t=-10, VIL data at time t=-15, etc.). When using VIL data at time t=-10, for example, a precipitation area that at least partially overlaps with the precipitation area detected based on the VIL data at time t=-5 (i.e., the previous precipitation area) is detected, and training is performed using this detected precipitation area as the training precipitation area.

[0097] Furthermore, in this embodiment, the VIL data (training data) used to train the machine learning model is, for example, VIL data from up to 3 hours prior to the time when the linear precipitation band occurred. Moreover, if no precipitation area overlapping at least partially with the previous precipitation area is detected based on the above-mentioned VIL data, training using that VIL data and VIL data from earlier times may be omitted.

[0098] Here, the machine learning model is described as learning precipitation areas that are connected to linear rainbands, but the same machine learning model may also learn precipitation areas that are not connected to linear rainbands. Precipitation areas that are not connected to linear rainbands are defined as precipitation areas where linear rainbands did not form nearby even after a period of time had passed, or precipitation areas that are clearly located or time-dependent from linear rainbands. In this case, the machine learning model only needs to learn features related to precipitation areas that are not connected to linear rainbands and a label (for example, label "0") that indicates that the precipitation area is not connected to a linear rainband.

[0099] When the above-described learning process is repeated, it becomes possible to generate a machine learning model that, for example, when given features related to a given precipitation area as input, outputs a label indicating whether or not a linear precipitation band will occur as a result of the development of that precipitation area.

[0100] Furthermore, if the configuration involves detecting (analytically extracting) the precipitation area of ​​interest as described above and learning its features (for example, rainfall statistics), it becomes possible to generate a highly accurate machine learning model even with less training data, compared to, for example, learning directly from VIL data (upper-air rainfall data) as shown in Figure 3.

[0101] Furthermore, the trained machine learning model only needs to be held in the detection unit 123, and the training of the machine learning model may be performed, for example, inside the information processing device 10 or outside the information processing device 10.

[0102] Furthermore, machine learning models are trained based on machine learning algorithms such as logistic regression, support vector machines, random forests, gradient boosting decision trees, or neural networks.

[0103] According to such a machine learning model, the detection unit 123 can determine whether or not the target precipitation area is a precipitation area connected to a linear precipitation band, based on the output from the machine learning model (e.g., a label) when features related to the target precipitation area, such as the rainfall statistics mentioned above, are input to the machine learning model.

[0104] Specifically, assuming the learning described above is performed, if the label "1" is output from the machine learning, it is determined that the target precipitation area is a precipitation area connected to a linear precipitation band, and the detection unit 123 can detect the occurrence of the linear precipitation band.

[0105] On the other hand, if the machine learning model outputs the label "0", it is determined that the target precipitation area is not a precipitation area connected to a linear precipitation band, and the detection unit 123 does not detect the occurrence of the linear precipitation band.

[0106] Here, the Japan Meteorological Agency issues a warning to areas where linear rainbands are likely to form about half a day in advance. However, the occurrence of linear rainbands in this case is predicted based on numerical weather prediction models used in weather forecasts, for example, so the accuracy of the prediction is low, and there are many cases where linear rainbands are missed or the prediction is wrong. Missing a linear rainband means that a linear rainband occurred, but its occurrence could not be predicted. A wrong prediction means that a linear rainband was predicted, but it did not occur.

[0107] In contrast, the configuration using the machine learning model described above can, for example, detect the occurrence of a linear rainband several hours before the Japan Meteorological Agency makes an announcement regarding its occurrence, and reduce missed occurrences and inaccurate predictions (i.e., improve detection accuracy).

[0108] Incidentally, although this explanation assumes the use of rainfall statistics as features related to the target precipitation area, there are still many unknowns regarding the occurrence of linear rainbands, and it is difficult to detect the occurrence of such linear rainbands with high accuracy using only a single feature (judgment index). For this reason, when detecting the occurrence of linear rainbands using a machine learning model as described above, at least two or more features should be used. These at least two or more features may be, for example, the maximum value, average value, or sum of the time-integrated VIL calculated as the rainfall statistics mentioned above, or they may be other than the rainfall statistics.

[0109] The following describes examples of features other than rainfall statistics that can be used when detecting the occurrence of linear rainbands using machine learning models.

[0110] First, the feature quantities may be calculated from wind condition data stored in the wind condition data storage unit 13, as explained in Figure 8. In this case, the feature quantities may include, for example, the wind condition statistics mentioned above. Alternatively, statistical values ​​(rainfall statistics or wind condition statistics) relating to the upwind region of the target precipitation area, or rainfall statistics based on multiple cumulative time periods may be used as feature quantities.

[0111] Furthermore, the features may be calculated from, for example, the prediction data stored in the prediction data storage unit 14 shown in Figure 11. In this case, for example, rainfall statistics calculated from the prediction data (predicted distribution of rainfall) can be used as features.

[0112] As shown in Figure 12, the feature quantities (rainfall statistics) in this case may be calculated from prediction data showing the predicted rainfall (e.g., VIL) at each of the multiple locations included in the target precipitation area from the present (time of detection processing) to 60 minutes later, or from VIL data showing the VIL (VIL calculated from the actually observed rainfall in the real space) at each of the multiple locations included in the target precipitation area from 30 minutes ago to the present, and prediction data showing the predicted rainfall (e.g., VIL) at each of the multiple locations included in the target precipitation area from the present to 30 minutes later.

[0113] In other words, the rainfall statistics used as features may be future rainfall statistics calculated solely from forecast data, or past and future rainfall statistics calculated from both the forecast data and VIL data. Furthermore, both rainfall statistics calculated from forecast data and rainfall statistics calculated from VIL data may be used as different features.

[0114] As described above, by using a machine learning model to detect the occurrence of linear rainbands, at least two of the following can be used as features: rainfall statistics for the target precipitation area, wind condition statistics for the target precipitation area, statistics for the upwind region of the target precipitation area, rainfall statistics based on multiple cumulative time periods, and rainfall statistics calculated from forecast data. This allows for high-precision detection of linear rainbands and improved lead time.

[0115] Furthermore, in order to improve the accuracy of detecting linear rainbands, it is necessary to select a combination of features from among the various features mentioned above that is useful for improving detection accuracy. However, it is difficult for users to perform manual analysis to select such a combination of features from among various features.

[0116] However, by generating a machine learning model that learns an arbitrary combination of features from among various features, and then evaluating the accuracy of that machine learning model, it becomes possible to easily select (determine) a useful combination of features compared to the manual analysis described above.

[0117] Furthermore, there are multiple factors that contribute to the formation of linear rainbands, making it difficult to detect all occurrences of linear rainbands with a single model. Additionally, there may be variations in lead time and detection accuracy among different machine learning models. Therefore, it is advisable to have multiple machine learning models (hereinafter referred to as the first and second machine learning models) trained with different combinations of features. With such a configuration, the final detection result (i.e., whether or not a linear rainband will occur) can be determined based on, for example, the result of determining whether a target precipitation area is connected to a linear rainband (hereinafter referred to as the "determination result by the first machine learning model") obtained by inputting features into the first machine learning model, and the result of determining whether a target precipitation area is connected to a linear rainband (hereinafter referred to as the "determination result by the second machine learning model") obtained by inputting different features into the second machine learning model.

[0118] In this case, the final detection result can be determined, for example, by a logical AND or OR operation on the determination results of the first and second machine learning models. Specifically, in a configuration where the final determination result is determined by a logical AND operation on the determination results of the first and second machine learning models, the occurrence of a linear rainband will be detected if both the first and second machine learning models determine that the target precipitation area is a precipitation area connected to a linear rainband. Also, in a configuration where the final detection result is determined by a logical OR operation on the determination results of the first and second machine learning models, the occurrence of a linear rainband will be detected if at least one of the first and second machine learning models determines that the target precipitation area is a precipitation area connected to a linear rainband. The final detection result may also be determined by considering the weighting of each of the determination results of the first and second machine learning models.

[0119] Figure 13 shows an example of a combination of features input to the first machine learning model, and Figure 14 shows an example of a combination of features input to the second machine learning model. However, since the first and second machine learning models use different features, the results of their judgments are likely to show different trends. Therefore, by comprehensively considering the judgment results of the first and second machine learning models to detect the occurrence of linear precipitation bands, it may be possible to improve the detection accuracy.

[0120] Furthermore, generally speaking, there is a trade-off between the probability of a prediction being incorrect (the rate of incorrect predictions) and the lead time. Specifically, determining the final detection result by performing a logical AND operation on the judgment results of the first and second machine learning models described above shortens the lead time but reduces the rate of incorrect predictions. On the other hand, determining the final detection result by performing a logical OR operation on the judgment results of the first and second machine learning models increases the rate of incorrect predictions but lengthens the lead time.

[0121] Therefore, in the configuration using the first and second machine learning models as described above, the method for determining the final detection result may be changed depending on whether the false alarm rate or lead time is prioritized. Furthermore, the configuration may be such that the machine learning model (or combination of machine learning models) used to detect the occurrence of linear rainbands is selected depending on whether the false alarm rate or lead time is prioritized. Note that the method for determining the final detection result and the selection of the machine learning model used to detect the occurrence of linear rainbands may be done, for example, depending on the type of linear rainband or the user's needs.

[0122] Furthermore, while this explanation primarily focuses on the use of two machine learning models (the first and second machine learning models), configurations using three or more machine learning models are also acceptable. In cases where three or more machine learning models are used, the final detection result may be determined by a majority vote of the judgment results from those models.

[0123] Furthermore, while we have described multiple machine learning models with different input features, these multiple machine learning models may differ not in their input features, but in the machine learning algorithms used to generate them.

[0124] Although several features that can be used to detect the occurrence of linear rainbands have been described, the configuration of the information processing device 10 may be changed depending on the combination of features used. Specifically, when VIL data, wind condition data, and forecast data are used to calculate the features, the information processing device 10 can be configured by combining the configurations shown in Figures 8 and 11, and including a rainfall data storage unit 11, a wind condition data storage unit 13, and a forecast data storage unit 14.

[0125] Furthermore, while this explanation describes detecting the occurrence of linear rainbands using a machine learning model that has learned the precipitation areas connected to linear rainbands, if the machine learning model learns precipitation areas connected to other characteristic precipitation areas (such as dangerous precipitation areas that do not fall under the category of linear rainbands but are accompanied by heavy rainfall that can cause damage), it becomes possible to use the machine learning model to detect the occurrence of such dangerous precipitation areas. In other words, in this embodiment, it is also possible to use the machine learning model to detect, for example, uniquely defined characteristic precipitation areas other than linear rainbands.

[0126] Furthermore, although a machine learning model has been described here, any other determination model besides the machine learning model may be used, as long as it is possible to determine whether or not the target precipitation area is connected to a linear precipitation band by inputting the features (determination indicators) related to the target precipitation area described above.

[0127] Incidentally, in this embodiment and its modifications, the information processing device 10 has been described as being located outside the weather radar system and acquiring three-dimensional rainfall data from the weather radar system. However, the information processing device 10 may also be mounted on the weather radar system.

[0128] Figure 15 shows an example of the configuration of a weather radar system equipped with an information processing device 10. As shown in Figure 15, the weather radar system 300 comprises a transceiver 301, a signal processing device 302, and an information processing device 10.

[0129] The transceiver 301 is configured to transmit a radar signal (radar wave) and receive a reflected wave signal based on the reflected wave of the radar signal. The transceiver 301 outputs the received reflected signal to the signal processing device 302.

[0130] The signal processing device 302 processes the reflected wave signal output from the transceiver 301 to acquire observation data, including, for example, the received power (echo intensity) of the reflected wave signal, and generates three-dimensional rainfall data based on the acquired observation data. The signal processing device 302 outputs the generated three-dimensional rainfall data to the information processing device 10.

[0131] The transceiver 301 and signal processing device 302 described above constitute a weather radar, such as a phased array weather radar (MP-PAWR or PAWR).

[0132] The three-dimensional rainfall data output from the signal processing device 302 is stored in the rainfall data storage unit 11 included in the information processing device 10 and used to detect the occurrence of linear rainbands, as described above.

[0133] This description explains the case where the information processing device 10 is installed in a weather radar system. However, in environments where multiple weather radar systems are deployed, for example, the observation data acquired by each of the weather radar systems (signal processing devices) may be synthesized and processed in a radar analysis and synthesis station. The information processing device 10 according to this embodiment and its modified form may be installed (deployed) in such a radar analysis and synthesis station.

[0134] Figure 16 shows an example of the configuration of a radar analysis and synthesis processing station equipped with an information processing device 10. As shown in Figure 16, the radar analysis and synthesis processing station 400 is communicatively connected to multiple weather radar systems 500 and includes a data receiving device 401, a radar analysis and synthesis device 402, and an information processing device 10.

[0135] Each of the multiple weather radar systems 500 is equipped with a weather radar consisting of the above-described transmitting and receiving device and signal processing device, and is configured to transmit observation data acquired by the operation of the weather radar.

[0136] The data receiving device 401 is configured to receive observational data transmitted from each of the multiple weather radar systems 500. The data receiving device 401 outputs the received observational data to the radar analysis and synthesis device 402.

[0137] The radar analysis and synthesis device 402 generates three-dimensional rainfall data based on observation data output from the data receiving device 401. The three-dimensional rainfall data generated by the radar analysis and synthesis device 402 is data that shows a wide range of reliable rainfall, obtained, for example, by synthesizing observation data from multiple weather radar systems 500. The radar analysis and synthesis device 402 outputs the generated three-dimensional rainfall data to the information processing device 10.

[0138] The three-dimensional rainfall data output from the radar analysis and synthesis device 402 is stored in the rainfall data storage unit 11 included in the information processing device 10 and used to detect the occurrence of linear rainbands, as described above.

[0139] While several embodiments of the present invention have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These embodiments can be carried out in a variety of other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents. [Explanation of Symbols]

[0140] 10...Information processing device, 10a...CPU, 10b...Non-volatile memory, 10c...RAM, 10d...Communication device, 11...Rainfall data storage unit, 12...Processing unit, 13...Wind condition data storage unit, 14...Prediction data storage unit, 121...Precipitation area detection unit, 122...Statistical value calculation unit, 123...Detection unit, 300, 500...Weather radar system, 301...Transmitting and receiving device, 302...Signal processing device, 400...Radar analysis and synthesis processing station, 401...Data receiving device, 402...Radar analysis and synthesis device.

Claims

1. Based on upper-air rainfall data showing hourly rainfall above multiple locations, the precipitation area is detected based on the first hourly cumulative value of rainfall at each of those locations. The system includes a processing unit that detects the occurrence of characteristic precipitation areas based on statistical values ​​of rainfall calculated for the detected precipitation areas. Information processing device.

2. The information processing device according to claim 1, wherein the upper-air rainfall data is data representing the vertically integrated rainwater volume calculated based on three-dimensional rainfall data representing the amount of rain in three-dimensional space.

3. The above-mentioned upper-air rainfall data is generated using a phased array weather radar.

4. The information processing device according to claim 1, wherein the statistical value includes the maximum value, average value, or total value of the second-hour cumulative rainfall above each of a plurality of points included in the precipitation area.

5. The information processing apparatus according to claim 4, wherein the time used to calculate the second time cumulative value is different from the time used to calculate the first time cumulative value.

6. The information processing apparatus according to claim 1, wherein when the occurrence of the characteristic precipitation area is detected, the processing unit outputs region information indicating a part of the precipitation area based on upper-air rainfall data indicating the amount of rain in the air above each of the multiple points included in the precipitation area and wind condition data indicating the wind speed and wind direction at each of the multiple points included in the precipitation area.

7. The information processing apparatus according to claim 1, wherein the processing unit detects a first region and a second region different from the first region based on the first time cumulative value, and the distance between the first region and the second region is less than a predetermined value, and detects a precipitation area including the first region and the second region.

8. The information processing device according to claim 1, wherein the precipitation area is corrected to a predetermined shape.

9. A transmitting and receiving device that transmits a radar signal and receives a reflected wave signal based on the reflected wave of the radar signal, A signal processing device that processes the received reflected wave signal to acquire observation data and generates three-dimensional rainfall data indicating rainfall in three-dimensional space based on the acquired observation data, The information processing apparatus according to any one of claims 1 to 8 It is equipped with, The information processing device includes a storage unit for storing three-dimensional rainfall data generated by the signal processing device, The aforementioned upper-air rainfall data is generated based on the three-dimensional rainfall data stored in the storage unit. Weather radar system.

10. A synthesis device that generates three-dimensional rainfall data showing rainfall in three-dimensional space by combining observation data transmitted from each of multiple weather radar systems, The information processing apparatus according to any one of claims 1 to 8 above It is equipped with, The information processing device includes a storage unit for storing three-dimensional rainfall data generated by the synthesis device, The aforementioned upper-air rainfall data is generated based on the three-dimensional rainfall data stored in the storage unit. Analysis and synthesis processing department.

11. Based on upper-air rainfall data showing hourly rainfall above multiple locations, the precipitation area is detected based on the first hourly cumulative value of rainfall at each of those locations. Based on the rainfall statistics calculated for the detected precipitation area, the occurrence of a characteristic precipitation area is detected. method.

12. On the computer, The method involves detecting precipitation areas based on the first hourly cumulative rainfall value at each of the multiple locations, using upper-air rainfall data that shows hourly rainfall above multiple locations. Based on the rainfall statistics calculated for the detected precipitation area, the occurrence of a characteristic precipitation area is detected. A program to execute.

Images