JP6908947B1 - Real-time inundation inundation prediction system, real-time inland inundation inundation prediction device, real-time inland inundation inundation prediction method, real-time inland inundation inundation prediction program, computer-readable recording medium and stored equipment - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform inundation prediction in real time reflecting a prediction of rainfall amount. SOLUTION: An inundation depth distribution group preservation unit 21 for preserving an inundation depth distribution group from an inundation depth distribution map A0 in which no inundation has occurred to an assumed maximum inundation depth distribution map An, and an inundation depth distribution group. Of these, when the rainfall intensity of j [mm / hr] falls for k [hr] with the inundation depth distribution map at a predetermined time as the initial condition, the inundation depth distribution group A (inundation depth distribution) The pre-calculation graph storage unit 22 that saves the pre-calculation graph showing the correlation between the rainfall intensity and the inundation depth distribution map created from the rainfall intensity-specific database that shows the number corresponding to the inundation depth distribution, and the rainfall amount. Based on the observed rainfall acquired by the data acquisition unit 10 and the predicted rainfall, a future inundation depth distribution map can be obtained from a plurality of inundation depth distribution maps stored in the inundation depth distribution group preservation unit 21. The pre-calculation graph storage unit 22 includes an inundation depth prediction unit that selects based on the correspondence defined by the pre-calculation graph stored in the pre-calculation graph storage unit 22. [Selection diagram] Fig. 1

Description

The present invention relates to a real-time inland waters inundation prediction system, a real-time inland waters inundation prediction device, a real-time inland waters inundation prediction method, a real-time inland waters inundation prediction program, a computer-readable recording medium, and a stored device.

Super-large typhoons and guerrilla rainstorms, which are thought to be caused by global warming and abnormal weather in recent years, have occurred in various places, and there are concerns about inundation and flooding. Therefore, inundation maps and the like are provided for the purpose of providing safe evacuation routes. Especially in recent years, the use of big data has been called for in various fields, and as an effective example, research on real-time prediction of water level based on meteorological data is being promoted.

For example, a real-time dynamic inundation simulation system that predicts inundation based on the observed rainfall has been proposed (Patent Document 1). However, in this real-time dynamic flood simulation system, the prediction is based only on the amount of rainfall that has fallen in the past, and the amount of rainfall changes from the past, for example, the amount of rainfall becomes heavier or weaker. There was a problem that the error became large. On the other hand, it is possible to change the inundation forecast in consideration of the rainfall forecast, but in this case, the amount of calculation according to the complicated terrain for each region becomes enormous, and real-time inundation forecast is performed. However, if real-time inundation prediction is to be performed, a considerably high level of computing power is required, and there is a problem that the calculation cost rises.

Japanese Unexamined Patent Publication No. 2004-197554

The present invention has been made in view of such a background, and one of the purposes thereof is real-time inundation inundation inundation prediction which enables accurate inundation prediction reflecting the prediction of rainfall amount in real time. It is an object of the present invention to provide a system, a real-time inundation inundation prediction device, a real-time inland inundation inundation prediction method, a real-time inland waters inundation inundation prediction program, a computer-readable recording medium, and a stored device.

Means for Solving Problems and Effects of Invention

According to the real-time inundation inundation prediction system according to the first aspect of the present invention, it is assumed that uniform rainfall occurs in the predicted area, and inundation occurs in a certain order at each position included in the predicted area. A real-time inundation inundation prediction system that predicts inland inundation caused by rainfall in real time under the assumed conditions. Inundation that preserves the inundation depth distribution group from the inundation depth distribution map A0 where no inundation has occurred in the predicted area at the start time t0 to the inundation depth distribution map An at the time tn, which is the assumed maximum scale, at predetermined time intervals. Of the inundation depth distribution group preserved in the depth distribution group preservation section and the inundation depth distribution group preservation section, the rainfall intensity of j [mm / hr] with the inundation depth distribution map at a predetermined time as the initial condition. Rainfall intensity and inundation created from a database by rainfall intensity that indicates the number of inundation depth distribution in the inundation depth distribution group A (inundation depth distribution) when A pre-calculation graph storage unit that stores a pre-calculation graph showing the correlation of the depth distribution map, a rainfall data acquisition unit that acquires the observed rainfall and the predicted rainfall, and an observation acquired by the rainfall data acquisition unit. Based on the predicted rainfall and the predicted rainfall, the future inundation depth distribution map is stored in the pre-calculation graph storage unit from among the plurality of inundation depth distribution maps stored in the inundation depth distribution group storage unit. It is possible to include an inundation depth prediction unit that is selected based on the correspondence relationship defined in the pre-calculation graph, and a display unit that displays the inundation depth distribution map selected by the inundation depth prediction unit. With the above configuration, instead of creating a future inundation depth distribution map by calculating each time, a light load can be obtained by selecting and displaying from a set of a plurality of inundation depth distribution maps calculated in advance according to the correspondence. It is possible to display the inundation depth distribution map with, and it is possible to realize inundation inundation prediction in real time without improving the calculation performance.

Further, according to the real-time inundation inundation inundation prediction system according to the second aspect, in addition to the above configuration, the inundation depth prediction unit displays an inundation depth distribution map at a predetermined time as an initial condition at a time when there is no inundation. It can be the inundation depth distribution map A0 at t0.

Further, according to the real-time inundation inundation prediction system according to the third embodiment, in addition to any of the above configurations, the water level set at each of a plurality of water level observation points set in the prediction area is further determined. A water level gauge for detection, a plurality of inundation depth distribution maps A0 to An stored in the inundation depth distribution group storage unit, and a water level gauge in which each water level gauge installed at the plurality of water level observation points Cx reacts with each other. The reaction time tp is provided with a water level gauge inundation depth database storage unit that stores the associated water level gauge inundation depth database, and the inundation depth prediction unit obtains an inundation depth distribution map at a predetermined time as an initial condition. It can be specified from the water level gauge inundation depth database so that the corresponding inundation depth distribution map Ap at the water level gauge reaction time tp where any of the plurality of water level gauges reacts. With the above configuration, the initial condition of inundation inundation prediction is set to the inundation depth distribution map Ap corresponding to the water level observation point where inundation was detected, and more accurate inland inundation inundation prediction is performed based on the latest inundation situation. It becomes possible.

Furthermore, according to the real-time inundation inundation prediction system according to the fourth aspect, in addition to any of the above configurations, the inundation depth prediction unit was detected when a plurality of water level gauges reacted at the same time. The reaction of the water level gauge with the higher water level can be adopted to select the inundation depth distribution map. With the above configuration, by adopting a higher water level, it is possible to predict a higher water level and improve safety.

Furthermore, according to the real-time inundation inundation prediction system according to the fifth embodiment, in addition to any of the above configurations, the rainfall intensity-based database stored in the rainfall intensity-based database storage unit is displayed on the horizontal axis. It is possible to make a graph in which the elapsed time and the vertical axis are the amount of inundation.

Furthermore, according to the real-time inundation inundation inundation prediction system according to the sixth embodiment, in addition to any of the above configurations, the pre-calculation graph storage unit is inundated with an external rainfall force larger than the drainage capacity in the predicted area. It is possible to provide a rainfall intensity-specific database storage unit that stores the rainfall intensity-specific database for each of the increase period and the inundation decrease period in which the external rainfall force is smaller than the drainage capacity in the predicted area.

Furthermore, according to the real-time inundation inundation prediction system according to the seventh embodiment, in addition to any of the above configurations, the pre-calculation graph stored in the pre-calculation graph storage unit is used for the rainfall intensity. The vertical axis of another database can be a graph obtained by converting the inundation amount into the inundation depth distribution groups A0 to An.

Furthermore, according to the real-time inundation inundation prediction system according to the eighth embodiment, in addition to any of the above configurations, the pre-calculation graph storage unit further indicates the amount of rainfall in the rainfall intensity database on the horizontal axis. Includes an accuracy graph storage unit that stores the accuracy graph with the inundation depth distribution group A0-n taken in the order of the inundation depth distribution number and the vertical axis as the accuracy rate when the intensity j [mm / hr] continues for k hours. The pre-calculation graph stored in the pre-calculation graph storage unit is obtained from the accuracy graph stored in the accuracy graph storage unit, the vertical axis of the rainfall intensity-based database, and the inundation amount. It can be converted to the inundation depth distribution number in the inundation depth distribution map.

Furthermore, according to the real-time inundation inundation inundation prediction system according to the ninth aspect, in addition to any of the above configurations, an inundation depth distribution group creation unit for creating the inundation depth distribution group and the advance A pre-calculation graph creation unit for creating a calculation graph can be provided.

Furthermore, according to the real-time inundation inundation prediction device according to the tenth embodiment, it is assumed that uniform rainfall occurs in the predicted area, and inundation occurs in a certain order at each position included in the predicted area. A real-time inundation inundation predictor that predicts inland inundation caused by rainfall in real time under the assumed conditions. Inundation depth distribution group from the inundation depth distribution map A0 where no inundation has occurred in the predicted area at the start time t0 to the inundation depth distribution map An at the time tn, which is the maximum expected scale, is preserved at predetermined time intervals. Of the inundation depth distribution group preserved in the depth distribution group preservation section and the inundation depth distribution group preservation section, the rainfall intensity of j [mm / hr] with the inundation depth distribution map at a predetermined time as the initial condition. Rainfall intensity and inundation created from a database by rainfall intensity that indicates the number of inundation depth distribution in the inundation depth distribution group A (inundation depth distribution) when A pre-calculation graph storage unit that stores a pre-calculation graph showing the correlation of the depth distribution map, a database access unit that can access the database access unit, a rainfall data acquisition unit that acquires the observed rainfall and the predicted rainfall, and the above. Based on the observed rainfall acquired by the rainfall data acquisition unit and the predicted rainfall, a future inundation depth distribution map can be obtained from a plurality of inundation depth distribution maps stored in the inundation depth distribution group preservation unit. An inundation depth prediction unit that can be selected and output based on the correspondence defined by the pre-calculation graph stored in the pre-calculation graph storage unit can be provided. With the above configuration, instead of creating a future inundation depth distribution map by calculating each time, a light load can be obtained by selecting and displaying from a set of a plurality of inundation depth distribution maps calculated in advance according to the correspondence. It is possible to display the inundation depth distribution map with, and it is possible to realize inundation inundation prediction in real time without improving the calculation performance.

Furthermore, according to the real-time inundation inundation prediction method according to the eleventh form, it is assumed that uniform rainfall occurs in the predicted area, and inundation occurs in a certain order at each position included in the predicted area. It is a real-time inundation inundation prediction method that predicts inland inundation caused by rainfall in real time under the assumed conditions. From the inundation depth distribution map A0 where no inundation has occurred in the predicted area at the start time t0, an inundation depth distribution group up to the inundation depth distribution map An at the time tn, which is the assumed maximum scale, is created at predetermined time intervals. J [mm / hr] of the step of preserving in the inundation depth distribution group storage section and the inundation depth distribution map at a predetermined time among the inundation depth distribution groups stored in the inundation depth distribution group preservation section as initial conditions. It was created from a database by rainfall intensity that shows the number corresponding to the inundation depth distribution of the inundation depth distribution group A (inundation depth distribution) when the rainfall of rainfall intensity falls for k [hr]. , The process of creating a pre-calculation graph showing the correlation between rainfall intensity and the inundation depth distribution map and saving it in the pre-calculation graph storage unit, and the process of acquiring the observed rainfall and the predicted rainfall in the rainfall data acquisition unit. Based on the observed rainfall acquired by the rainfall data acquisition unit and the predicted rainfall, one of the plurality of inundation depth distribution maps stored in the inundation depth distribution group preservation unit will be selected in the future. As an inundation depth distribution map, a step selected by the inundation depth prediction unit and an inundation selected by the inundation depth prediction unit based on the correspondence defined by the pre-calculation graph stored in the pre-calculation graph storage unit. A step of displaying the depth distribution map on the display unit can be included. As a result, instead of creating a future inundation depth distribution map by calculating each time, it is possible to select and display from a set of a plurality of inundation depth distribution maps calculated in advance according to the correspondence relationship, with a light load. It is possible to display the inundation depth distribution map, and it is possible to realize inundation inundation prediction in real time without improving the calculation performance.

Furthermore, according to the real-time inundation inundation prediction program according to the twelfth form, it is assumed that uniform rainfall occurs in the predicted area, and inundation occurs in a certain order at each position included in the predicted area. A real-time inundation inundation prediction program that predicts inland inundation caused by rainfall in real time under the assumed conditions. From the inundation depth distribution map A0 where no inundation has occurred in the predicted area at the start time t0, an inundation depth distribution group up to the inundation depth distribution map An at the time tn, which is the assumed maximum scale, is created at predetermined time intervals. Of the inundation depth distribution group stored in the inundation depth distribution group storage section and the inundation depth distribution group stored in the inundation depth distribution group preservation section, j [mm / hr] of the inundation depth distribution map at a predetermined time is used as the initial condition. It was created from a database by rainfall intensity that shows the number corresponding to the inundation depth distribution of the inundation depth distribution group A (inundation depth distribution) when the rainfall of rainfall intensity falls for k [hr]. , A function to create a pre-calculation graph showing the correlation between rainfall intensity and inundation depth distribution map and save it in the pre-calculation graph storage section, and a function to acquire the observed rainfall and the predicted rainfall in the rainfall data acquisition section. Based on the observed rainfall acquired by the rainfall data acquisition unit and the predicted rainfall, one of the plurality of inundation depth distribution maps stored in the inundation depth distribution group preservation unit will be selected in the future. As an inundation depth distribution map, a function selected by the inundation depth prediction unit and an inundation selected by the inundation depth prediction unit based on the correspondence defined by the pre-calculation graph stored in the pre-calculation graph storage unit. The function of displaying the depth distribution map on the display unit can be realized in the computer. With the above configuration, instead of creating a future inundation depth distribution map by calculating each time, a light load can be obtained by selecting and displaying from a set of a plurality of inundation depth distribution maps calculated in advance according to the correspondence. It is possible to display the inundation depth distribution map with, and it is possible to realize inundation inundation prediction in real time without improving the calculation performance.

Furthermore, the computer-readable recording medium or storage device according to the thirteenth embodiment stores the above program. Recording media include CD-ROM, CD-R, CD-RW, flexible disc, magnetic tape, MO, DVD-ROM, DVD-RAM, DVD-R, DVD + R, DVD-RW, DVD + RW, and Blu-ray (registered). Includes magnetic disks such as (trademarks), HD DVDs (AODs), optical disks, magneto-optical disks, semiconductor memories, and other media capable of storing programs. Further, the program includes a program stored in the above-mentioned recording medium and distributed, and a program distributed by download through a network line such as the Internet. Further, the stored device includes a general-purpose or dedicated device in which the above program is implemented in a state in which it can be executed in the form of software, firmware, or the like. Furthermore, each process or function included in the program may be executed by computer-executable program software, or each part of the process may be performed by hardware such as a predetermined gate array (FPGA, ASIC), or the program software and hardware. It may be realized in a form in which a partial hardware module that realizes a part of the hardware is mixed.

This is a block showing a real-time inundation inundation prediction system according to the first embodiment. It is a block showing a real-time inundation inundation prediction system connected to a network. FIG. 3A is a schematic cross-sectional view showing the calculated ground height of the inundation analysis, and FIG. 3B is a schematic cross-sectional view showing the local ground height. It is a schematic diagram which shows the inundation depth distribution group A0 to An obtained by the basic unit calculation. It is a graph which shows the database by rainfall intensity in the inundation increase period. It is a graph which shows the database by rainfall intensity in the inundation decrease period. It is a graph which shows another example of the database by rainfall intensity in the inundation increase period. It is a graph which shows another example of the database by rainfall intensity in the inundation decrease period. It is a schematic diagram explaining the accuracy rate. It is a figure which shows an example of the accuracy rate graph in the inundation increase period. It is a figure which shows another example of the accuracy rate graph. It is a figure which shows an example of the pre-calculation graph in the inundation increase period. It is a figure which shows an example of the accuracy rate graph in the inundation decrease period. It is a figure which shows an example of the pre-calculation graph in the inundation reduction period. It is a flowchart which shows the procedure at the time of the actual operation of the real-time inundation inundation inundation prediction. It is a graph which shows the rainfall waveform from the beginning of the rain when the water level gauge is not used. It is a graph which shows another example of the pre-calculation graph in the inundation increase period. It is a graph which shows other example of the pre-calculation graph in the inundation reduction period. It is a schematic diagram which shows the relationship between the specified water level of a water level gauge and the water level of inundation depth distribution maps Ap-1 and Ap. It is a graph which shows the rainfall waveform from the beginning of the rain when using a water level gauge. It is a graph which shows another example of a rainfall waveform. 22A to 22E are schematic views showing how the display of the inundation depth distribution is switched in the real-time inundation prediction according to the first embodiment. 23A to 23B are schematic views showing how the display of the inundation depth distribution is switched in the real-time inundation prediction according to the second embodiment. It is a graph which shows the example of the updated rainfall waveform. 25A to 25B are schematic views showing how the display of the inundation depth distribution is switched under the updated conditions in the real-time inundation prediction according to the second embodiment. It is a schematic diagram which shows the target area by the inland waters inundation inundation prediction method which concerns on Embodiment 3. It is a graph which shows the rainfall waveform given as the external force of the basic unit calculation. It is a graph which shows the amount of flooding when the rainfall waveform of FIG. 27 is given. It is a graph which shows the water level when the rainfall waveform of FIG. 27 is given. It is a graph which shows the time change of the water level in the water level increase period. It is a graph which shows the time change of the water level in the water level decrease period. It is a graph which shows the time change of the inundation depth distribution group in the water level increase period. It is a graph which shows the time change of the inundation depth distribution group in the water level decrease period. It is a graph which shows the rainfall waveform given as an external force. It is a graph which shows the time change of the inundation depth distribution group output when the water level gauge is not used. It is a graph which shows the time change of the inundation depth distribution group output when using a water level gauge. It is an image diagram which shows an example of the inundation depth distribution map displayed on the display part. It is a graph which plotted the inundation depth of the inundation depth distribution map at the same place. It is a graph which plotted the difference of the inundation depth of the inundation depth distribution map at the same place.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, the embodiments shown below are examples for embodying the technical idea of the present invention, and the present invention is not specified as the following. Further, the present specification does not specify the members shown in the claims as the members of the embodiment. In particular, the dimensions, materials, shapes, relative arrangements, etc. of the components described in the embodiments are not intended to limit the scope of the present invention to the specific description, but are merely described. It's just an example. The size and positional relationship of the members shown in each drawing may be exaggerated to clarify the explanation. Further, in the following description, members having the same or the same quality are shown with the same name and reference numeral, and detailed description thereof will be omitted as appropriate. Further, each element constituting the present invention may be configured such that a plurality of elements are composed of the same member and the plurality of elements are combined with one member, or conversely, the function of one member is performed by the plurality of members. It can also be shared and realized.

The connection between the real-time inundation inundation prediction device used in the embodiment of the present invention and the computer, printer, external storage device and other peripheral devices connected to the real-time inundation inundation prediction device for operations, control, display, other processing, etc. For example, serial connection such as IEEE1394, RS-232x, RS-422, RS-423, RS-485, USB, parallel connection, or electrical via a network such as 10BASE-T, 100BASE-TX, 1000BASE-T, etc. Alternatively, they are magnetically and optically connected for communication. The connection is not limited to a physical connection using a wire, and may be a wireless LAN such as IEEE 802.1x, a wireless connection using Bluetooth (registered trademark), other radio waves such as NFC, infrared rays, optical communication, or the like. Further, a memory card, a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like can be used as a recording medium for exchanging data, saving settings, and the like. In the present specification, the real-time inundation inundation prediction device means not only the real-time inland inundation inundation prediction device itself but also an inland inundation inundation prediction system in which peripheral devices such as a computer and an external storage device are combined. Used in.
[Embodiment 1]

The real-time inundation inundation prediction system predicts the inundation of inland waters in the prediction area for which the inundation prediction is performed. FIG. 1 shows a real-time inundation inundation prediction system 1000 according to the first embodiment of the present invention. The real-time inundation inundation prediction system 1000 shown in this figure predicts the inundation depth distribution map of the predicted area based on the rainfall data in a certain predicted area, and displays it on the display unit 40. Here, the predicted area refers to the entire area where the presence or absence of inundation, the degree of inundation risk, etc. are predicted by the real-time inundation inundation inundation prediction system 1000. The range of this forecast area is not specified. The range of the predicted area can be arbitrarily determined by the system user within the target area, and can be set to a specific municipality unit, for example. In addition, the inundation depth distribution map is a map that is superimposed on the map to display the locations where inundation is occurring and the water depths thereof.

The real-time inundation inundation prediction system 1000 includes a real-time inland inundation inundation prediction device 100. The real-time inundation inundation prediction device 100 can be configured by installing the inland inundation inundation prediction program on a general-purpose or dedicated computer in addition to being configured by dedicated hardware. The real-time inundation inundation prediction system 1000 includes a rainfall data acquisition unit 10, a data storage unit 20, a calculation unit 30, a display unit 40, and an operation unit 50.
(Rainfall data acquisition unit 10)

The rainfall data acquisition unit 10 is a member for acquiring data from the outside. Here, it functions as a member for acquiring rainfall data. Rainfall data is typically rainfall per unit time. The rainfall includes the observed rainfall observed in the past (past rainfall) and the predicted rainfall predicted future rainfall (future rainfall). Rainfall data is acquired via the network. This makes it easy to update the latest information one by one. Therefore, the rainfall data acquisition unit 10 is provided with a communication function for connecting to a specific network via a general-purpose network line such as the Internet or a dedicated line.

Rainfall data includes analytical rainfall and short-term precipitation forecasts for the forecast area. Here, the analyzed rainfall is the amount of rainfall actually rained within a predetermined time in the past from the current time. The short-term precipitation forecast is the amount of rainfall that is expected to rain within a predetermined time from the current time. In the present embodiment, this predetermined time is set to every hour. However, it may be any time unit such as every 30 minutes.

For example, the rainfall data distributed by the Meteorological Business Support Center is distributed by updating the analyzed rainfall and short-term precipitation forecast every 30 minutes with the predicted rainfall up to 6 hours ahead in the range where the topographical data is divided by a 1 km mesh. Therefore, the rainfall data acquisition unit 10 sequentially acquires such data and sends it to the calculation unit 30. Further, the rainfall data captured by the rainfall data acquisition unit 10 can be stored in the data storage unit 20. For example, the data storage unit 20 may be provided with a rainfall data storage unit 21 for holding rainfall data.

The collection destination of the rainfall data is not particularly specified. For example, the rainfall data distributed by the Japan Meteorological Agency or the Meteorological Business Support Center may be used, or an original rainfall observation device or the like may be installed to directly collect the rainfall data. Further, the predetermined time is not limited to one hour, and may be a shorter time (for example, 30 minutes) or a longer time (for example, 2 hours).
(Data storage unit 20)

The data storage unit 20 is a member for holding various data. The data storage unit 20 includes an inundation depth distribution group storage unit 21 and a pre-calculation graph storage unit 22.

The inundation depth distribution group preservation unit 21 preserves the inundation depth distribution group. The inundation depth distribution group is a set of inundation depth distribution maps Ai showing the inundation status of the predicted area at a specific time ti, and is predetermined from the inundation depth distribution map A0 in which inundation does not occur in the predicted area at the inundation start time t0. It is a set up to the inundation depth distribution map An at the time tn, which is the assumed maximum scale in time increments. This inundation depth distribution group is generated in advance by the inundation depth distribution group creation unit 33 and stored in the inundation depth distribution group preservation unit 21.

The pre-calculation graph storage unit 22 is a member for storing the pre-calculation graph. In the pre-calculation graph, among the inundation depth distribution groups stored in the inundation depth distribution group preservation unit 21, the rainfall intensity of j [mm / hr] is k, with the inundation depth distribution map at a predetermined time as the initial condition. Rainfall intensity and inundation depth distribution created from a database by rainfall intensity that indicates the number of inundation depth distribution in the inundation depth distribution group A (inundation depth distribution) when it falls between [hr]. It is a graph which shows the correlation of the figure. This pre-calculation graph is generated in advance by the pre-calculation graph creation unit 34 and stored in the pre-calculation graph storage unit 22.

The pre-calculation graph storage unit 22 includes a database storage unit 23 for each rainfall intensity and a predictive value graph storage unit 24.

The rainfall intensity-based database storage unit 23 stores the rainfall intensity-based database. The database by rainfall intensity is a graph in which the horizontal axis is the elapsed time and the vertical axis is the amount of inundation. In addition, the database by rainfall intensity is created for each of the inundation increase period when the external rainfall force is larger than the drainage capacity in the predicted area and the inundation decrease period when the external rainfall force is smaller than the drainage capacity in the predicted area, and stored in the database by rainfall intensity. Has been done.

The accuracy rate graph storage unit 24 stores the accuracy rate graph. In the predictive value graph, the horizontal axis is the inundation depth distribution group A0-n when the rainfall intensity of rainfall j [mm / hr] continues for k hours in the database by rainfall intensity, and the vertical axis is the inundation depth distribution number. It is a graph with the accuracy rate.

The pre-calculation graph is a graph obtained by converting the vertical axis of the database by rainfall intensity from the inundation amount to the inundation depth distribution groups A0 to An, that is, the inundation depth distribution number of the inundation depth distribution map.

Further, the data storage unit 20 may include a water level gauge inundation depth database storage unit 25. The water level gauge inundation depth database storage unit 25 stores the water level gauge inundation depth database. The water level gauge inundation depth database is a water level gauge reaction time in which a plurality of inundation depth distribution maps A0 to An stored in the inundation depth distribution group storage unit 21 and each water level gauge installed at a plurality of water level observation points Cx react respectively. It is an inundation depth database associated with tp (or the number of minutes in Ai).
(Calculation unit 30)

The calculation unit 30 is a member for performing various calculations, and here, fulfills the function of the inundation depth prediction unit 31. The inundation depth prediction unit 31 draws a future inundation depth distribution map from a plurality of inundation depth distribution maps based on the observed rainfall acquired by the rainfall data acquisition unit 10 and the predicted rainfall amount. It is a member for selection based on the correspondence specified in.

Further, the calculation unit 30 includes a database access unit 32 for accessing the data storage unit 20. The database access unit 32 is an interface for accessing the inundation depth distribution group storage unit 21, the pre-calculation graph storage unit 22, and the like. Further, the database access unit 32 has a configuration for accessing a storage medium such as a hard disk or a semiconductor memory included in the real-time inundation inundation prediction device 100, and also accesses a database connected to the outside of the real-time inland inundation inundation prediction device 100. It may be configured to be used. For example, by using an external storage such as a NAS or a file server connected through a network line as the data storage unit 20, a real-time inundation inundation prediction device 100 and a real-time inland inundation inundation prediction system 1000 can be constructed.

Further, as shown in the real-time inundation inundation inundation prediction system 1000'in FIG. 2, the calculation unit 30 includes an inundation depth distribution group creation unit 33 for creating an inundation depth distribution group and a pre-calculation graph creation unit for creating a pre-calculation graph. It can also perform 34 functions. However, since it is sufficient to create the inundation depth distribution group and the pre-calculation graph in advance and save them in the data recording unit 20, it is not always necessary to create them with the real-time inundation inundation inundation prediction device 100. For example, it suffices to create it with an arithmetic unit such as an external computer or server and register it in the data storage unit 20 or another database, respectively. In this case, it is not necessary to provide the inundation depth distribution group creation unit 33 and the pre-calculation graph creation unit 34 in the real-time inundation inundation inundation prediction device 100. FIG. 1 shows such a configuration example.

For such an inundation depth prediction unit 31, a CPU can be used, for example. Further, not limited to the CPU, a graphics processing unit (GPU) may be used. By using such GPGPU (General-Purpose computing on Graphics Processing Units), the load on the CPU is reduced, the calculation speed is improved, and the prediction result is made more accurate, while the display unit described later is used. 40 allows the system user to easily confirm the prediction result. For GPGPU, for example, CUDA (trade name) can be used.
(Display unit 40)

The display unit 40 is a member for displaying an inundation depth distribution map or the like selected by the inundation depth prediction unit. The display unit 40 can display a future inundation area prediction and a past inundation depth distribution. A liquid crystal display, an organic EL, a CRT, or other display can be used for such a display unit 40. Further, the display unit and the input unit may be used as the touch panel. FIG. 37 shows an example of the inundation depth distribution map displayed on the display unit 40.

The display unit 40 has a map display area 41 for displaying a map of the predicted area. In the map display area 41, the map is displayed at a predetermined magnification. The magnification of the map can be enlarged or reduced by operating the enlargement / reduction buttons provided at the corners of the map display area 41. Alternatively, the scroll buttons of the mouse constituting the operation unit 50 may be operated to enlarge or reduce the size. In addition, for data management, grid-like lines are displayed vertically and horizontally with a predetermined width, and are displayed in a block shape divided by lines. Hereinafter, the area divided by the line is referred to as a block (reference area mesh). The map display magnification can be changed as appropriate.
(Operation unit 50)

The operation unit 50 is a member for performing various operations, and is composed of, for example, a pointing device such as a keyboard and a mouse, a console, and the like. Further, by using the touch panel, the operation unit 50 and the display unit 40 can be shared.
(Server / Client)

Further, by using the calculation unit as the server device and the display unit 40 as the display screen of the client device connected to the server device, the user can access the server device from the outside using the client device and obtain the result by the calculation unit. You can browse the prediction results and so on. An example of this is shown in FIG. In this case, the operation unit 50 can operate the client device CL connected to the server device SV via the network, for example, a touch panel, a mouse, or the like.

The display unit 40 displays the user interface screen of the inundation inundation prediction program. As described above, the user interface screen of the inundation inundation prediction program accesses the display unit 40 connected to the stand-alone computer and the server device in which the inland inundation inundation prediction program is installed as shown in FIG. The display screen of the client device is also included. By making the drawing web-based, it is possible to have versatility. As the client device, a smartphone, a mobile phone, a personal computer or the like having a communication function for accessing the server device can be used.

With such a configuration, the future inundation depth distribution map is not calculated and created each time, but is selected and displayed according to the correspondence from a set of a plurality of inundation depth distribution maps calculated in advance. It is possible to display the inundation depth distribution map with a light load, and it is possible to realize inundation inundation prediction in real time without improving the calculation performance. In particular, in the past, future forecasts were made based on past data such as observed rainfall. The error becomes large due to changes such as heavy rainfall or subsidence.
(Water level gauge WG)

If a water level gauge is installed in the prediction area where the real-time inundation inundation prediction system is used to predict inundation, the data of this water level gauge can be used to improve the accuracy of real-time inundation inundation prediction. .. In the examples of FIGS. 1 and 2, a plurality of water level gauges WG are installed in the predicted area.

The water level gauge WG is a device for observing the water level, which is also called a submersion sensor or the like. Further, the water level gauge WG may be a device for measuring how much the water level is as a numerical value, or a device for detecting that the water level has reached a preset predetermined water level. For example, a submergence sensor set to set the specified water level to a water level determined to be submerged and to operate when the water level reaches the specified water level can be used as a water level gauge WG.

Such water level gauges WG are installed at a plurality of locations in the prediction area where inundation prediction is performed. For example, roads such as highways in urban areas, designated evacuation sites, public facilities such as city halls, ponds, rivers, etc. can be mentioned. The place where the water level gauge WG is installed is called a water level observation point.

The water level gauge WG outputs a water level signal indicating the water level to the inundation prediction device. The water level gauge WG is connected to the inundation predictor by wire or wirelessly. As shown in FIG. 1, each water level gauge WG may be directly connected to the inundation prediction device or may be connected via a repeater. In the example of FIG. 2, the plurality of water level gauges WG are wirelessly connected to the inundation prediction device via the repeater RT. For example, a plurality of water level gauges WG and repeater RT are connected by a standardized communication method such as Bluetooth such as WiFi or BLE and ZigBee (both are product or service names). They are connected by public lines such as telephone lines and optical fibers.
(Specified water level of water level gauge WG)

An example of setting a specified water level with a water level gauge WG installed at a water level observation point will be described with reference to FIGS. 3A and 3B. In these figures, FIG. 3A is a schematic cross-sectional view showing the calculated ground height, that is, the ground surface and the water level in the calculated mesh, and FIG. 3B is the local ground height, that is, the water level detected by the water level gauge WG at the actual water level observation point. A schematic cross-sectional view showing each of the above is shown. Here, as an example of the installation altitude, T.I. P. (Tokyo Peil: Tokyo Bay average sea level, which is the origin of Japan's Geodetic Standard) Consider an example of installing a water level gauge WG at 1.4 m. Here, the calculated ground height of the inundation analysis is as shown in FIG. 3A. P. Although it is 1.2 m, the local ground height is T.I. P. It may be 1.0 m. In this case, by installing or setting a water level gauge WG so as to detect a position 0.4 m higher than the local ground height, the arrival time when the water depth reaches 20 cm in the flood analysis can be determined at this water level observation point. It can be the arrival time. That is, as shown in FIG. 3A, the altitude (TP) of the inundation analysis calculation mesh is 1.2 m, while the altitude (TP) of the actual installation point is as shown in FIG. 3B. In the case of 1.0 m, the specified water level of the water level gauge WG installed at this water level observation point is (TP) 1.4 m.

The real-time inundation inundation prediction method according to the first embodiment of the present invention includes an inundation analysis phase and a statistical analysis phase. The flood analysis phase includes calculation of basic units and creation of a database by rainfall intensity. In addition, as the statistical analysis phase, identification of the accuracy rate and identification of the inundation depth distribution map during actual operation can be mentioned. With this real-time inundation prediction method, it is possible to estimate the current inundation area and predict the future inundation area.
(Current inundation area estimation)

The current inundation area is estimated based on the observed rainfall and the observed water level of the water level gauge. The inundation area is predicted by selecting the one with the highest correlation from the simulation results (inundation depth distribution) of the inundation analysis.

Current inundation area estimation includes a database creation process for pre-inundation analysis and a process for identifying the predictive value. In addition, if a water level gauge is installed in the area, the current inundation area estimation further includes the step of specifying the inundation start time of the basic unit calculation at the water level observation point.

In the flood analysis, when the mesh in the area is flooded, the time from the start of rain until the water level reaches the specified water level is defined as the arrival time. Calculate the arrival time at the water level observation point by flood analysis in advance.

By performing these pre-inundation analysis database creation process (basic unit calculation process and rainfall intensity database creation process), water level gauge inundation depth database creation process, and predictive value identification process in advance, In actual operation, it is possible to easily predict the inundation area in the future by taking these calculation results into consideration, and it is possible to predict the inundation inundation in real time by reducing the calculation load.
(Forecast of future inundation area)

Predict future inundation areas based on the current inundation area estimates obtained. Here, the increase / decrease in the inundation area is predicted from the current water level obtained by the above method, using the future prediction of rainfall obtained from the Japan Meteorological Agency as an external force. Similar to the current inundation area prediction, the future inundation area prediction is performed by selecting the one with the highest correlation from the simulation results of the inundation analysis calculated in advance.

The details of the real-time inundation inundation prediction system will be described below.
(Create database by prior flood analysis)

The database creation process by the preliminary flood analysis includes the basic unit calculation process and the database creation by rainfall intensity.
(Basic unit calculation)

Inland inundation is carried out on the assumption that it will rain spatially uniformly in the area targeted for real-time inland inundation inundation prediction, and that the inundation order will be the same from the same location. Under these assumptions, inundation analysis is performed to create inundation depth distribution maps from the start of inundation to the assumed maximum scale. In the flood analysis here, the external rainfall force is created according to the past guidelines. The existing guideline is the Ministry of Land, Infrastructure, Transport and Tourism Water and Disaster Management Bureau "Method for setting the assumed maximum external force for creating inundation assumptions (floods, inland waters)" (July 2015). As an example, according to this, it is assumed that the existing guideline is the assumed maximum external force, and the inundation analysis of the external force is performed by extending the existing rainfall to the assumed maximum rainfall in the area where the inundation is predicted. This is based on the premise that no more inundation than calculated here will occur. For example, the assumed maximum rainfall is the past rainfall, that is, the amount of rainfall that has occurred in the past multiplied by 1.3.

In this way, as shown in FIG. 4, an inundation depth distribution map A from the start of inundation to the assumed maximum scale is created every minute. In the inundation depth distribution map A0 at the initial value of 0 minutes, it is assumed that no inundation has occurred. The set of the inundation depth distribution maps A0 to An obtained in this way from the 0th minute to the nth minute is called an inundation depth distribution group (group A). The process of calculating the inundation depth distribution group is called the basic unit calculation, and the obtained result is called the basic unit calculation result. By such basic unit calculation, the forecast of the maximum damaged area and the database of how much area is inundated with respect to the rainfall time are created in advance.

In summary, the inundation depth distribution group is a set of inundation depth distribution maps Ai showing the inundation status of the predicted area at a specific time ti. This inundation depth distribution group includes 0 to n sheets from the inundation depth distribution map A0 in which inundation does not occur in the predicted area at the inundation start time t0 to the inundation depth distribution map An at the assumed maximum scale time tun. .. In this example, an example of creating an inundation depth distribution map every minute has been described, but the present invention does not limit the time width for creating an inundation depth distribution map to 1 minute, for example, every 5 minutes or every 30 minutes. It can be any time width, such as every hour. The shorter the time width, the more accurate real-time inundation and inundation prediction becomes possible. However, the amount of calculation is also large. The inundation depth distribution group created in this way is stored in the inundation depth distribution group preservation unit 21.
(Create database by rainfall intensity)

Next, a database for each rainfall intensity is created and stored in the database storage unit 23 for each rainfall intensity. The purpose of creating a database by rainfall intensity is to determine which, that is, what number of inundation depth distribution in the inundation depth distribution group, when the rainfall intensity of j [mm / hr] falls in the time width of k [hr]. Whether it corresponds to the figure or not, it is to identify the correspondence. Here, the rainfall intensity j is set as a plurality of different values and calculated for each. Preferably, the lower limit of the amount of rainfall is set to the amount of rainfall in which inundation does not occur unless the amount of rainfall is higher than that in the predicted area where the inundation is predicted. The upper limit is the maximum expected rainfall. Further, the initial condition is an inundation depth distribution corresponding to the inundation depth distribution map Ai. (Details will be described later)

The database by rainfall intensity is created separately for the inundation increase period and the inundation decrease period. The graphs of the obtained database by rainfall intensity are shown in FIGS. 5 and 6. As shown in these figures, the horizontal axis represents the elapsed time k [hr] and the vertical axis represents the amount of flooding.

In the period of increased inundation, the external rainfall force is defined as a rainfall waveform in which rainfall with a rainfall intensity of j [mm / hr] continues. However, j> dr (dr: drainage capacity [mm / hr]). The initial condition is the inundation depth distribution map A0 in the state where there is no inundation. The database by rainfall intensity during the inundation increase period thus obtained is as shown in FIG. 5, and shows a tendency that the inundation depth distribution increases with the elapsed time.

On the other hand, in the preparation of the database by rainfall intensity during the inundation reduction period, the rainfall external force is also set to the rainfall waveform in which the rainfall intensity j [mm / hr] continues to fall. However, here, j <dr (dr: drainage capacity [mm / hr]). The initial condition is the assumed maximum scale inundation depth distribution map An. The database by rainfall intensity during the inundation reduction period thus obtained shows that the inundation depth distribution tends to decrease with the elapsed time as shown in FIG.

In the examples of FIGS. 5 and 6, the vertical axis of the database for each rainfall intensity is the inundation amount, but other parameters such as the size of the inundation area (the number of meshes) may be used. Such an example is shown in FIGS. 7 and 8. In these figures, FIG. 7 shows the calculation results when 15 mm / hr, 20 mm / hr, ..., The assumed maximum rainfall [mm / hr] continues to fall. Here, since the drainage capacity dr is set to 10 mm / hr in the predicted area where the inundation is predicted, it is assumed that inundation does not occur unless the rainfall exceeds 10 mm / hr. In addition, FIG. 8 shows the results of inundation analysis using the rainfall waveform as an external force when the inundation depth distribution of the assumed maximum scale heavy rainfall is set as the initial condition and the rainfall of 10 mm / hr or less continues to fall. Here, since it is 10 mm / hr or less, the drainage capacity of the area exceeds the rainfall, indicating that the inundation decreases with time. Although the graphs are shown as straight lines in FIGS. 7 and 8 for convenience, it is assumed that the actual graph does not become a straight line.
(Correlation between rainfall intensity and basic unit calculation)

Next, based on the basic unit calculation obtained in the flood analysis phase and the database by rainfall intensity, the correlation between the rainfall intensity and the basic unit calculation is obtained as the statistical analysis phase. Specifically, the correspondence between the elapsed time of rainfall and the inundation depth distribution map is obtained. More specifically, the vertical axis of the database for each rainfall intensity in FIGS. 5 and 6 is converted from the inundation amount to the inundation depth distribution number (0 to n) indicating the number of the inundation depth distribution map A.

Here, the accuracy rate is first calculated, and then the maximum value of this accuracy rate is calculated in the inundation increase period and the inundation decrease period, respectively. Furthermore, the vertical axis of the database creation by rainfall intensity is replaced with the time when the maximum correlation was obtained in the basic unit calculation, that is, the minute in which the maximum correlation was obtained in the basic unit calculation.
(Calculation of accuracy rate)

（適中率グラフの作成） First, the accuracy rate is calculated. Here, it is calculated how many minutes the inundation area of the basic unit calculation result has the highest correlation with the inundation area for each rainfall intensity obtained in the process of creating the database for each rainfall intensity. This situation is shown in FIG. In FIG. 9, in the horizontal direction, inundation depth distribution when rainfall of rainfall intensity j [mm / hr] continues for k hours in the database by rainfall intensity is taken as with or without inundation. Further, in the vertical direction, in the inundation depth distribution map Ai obtained by the basic unit calculation, the presence or absence of inundation is taken. From this matrix, the predictive value of Equation 1 is obtained as an expression showing the correlation. This accuracy rate is converted by the number of meshes. The predictive value is calculated for each rainfall.
[Number 1]

(Creation of accuracy graph)

Next, the maximum value of the predictive value obtained in this way is obtained for each rainfall amount. In addition, the maximum value of such an accuracy rate is calculated in each of the inundation increase period and the inundation decrease period. First, the maximum value of the predictive value in the inundation increase period is calculated.
(Maximum predictive value during the period of increased inundation)

Specifically, as shown in FIG. 10, on the horizontal axis, the inundation depth distribution groups A0-n when the rainfall intensity j [mm / hr] continues for k hours in the database by rainfall intensity are arranged in the order of the inundation depth distribution number. Therefore, a predictive value graph is created with the vertical axis as the predictive value. This predictive value graph is created for each rainfall.

As an example, the predictive value graph of FIG. 11 shows the predictive value when it rains continuously with a rainfall of 20 mm / hr in the database by rainfall intensity. In this case, it is shown that the calculation result (inundation area) at the 10th minute has a high correlation with the inundation area at the 5th minute of the basic unit calculation result. , The inundation area of the basic unit calculation result with the highest predictive value is selected.

Next, from the predictive value graph, the vertical axis of the database by rainfall intensity is converted from the inundation amount to the inundation depth distribution number (any of the group A) of the inundation depth distribution map. Since it is assumed that inland inundation will inundate from the same location in the same order, the ● part in the database by rainfall intensity in Fig. 5 is one of Group A (inundation depth from the start of inundation to the assumed maximum scale). Equivalent to. In this way, the graph of FIG. 12 is obtained. The graph obtained by converting the vertical axis is called a pre-calculation graph. Such a pre-calculation graph is created in advance for each rainfall and registered in the pre-calculation graph database. In this example, the vertical axis of the pre-calculation graph is the inundation depth distribution group A0-n, but the present invention does not limit the vertical axis to the inundation depth distribution group, for example, the inundation depth distribution number 0 to n. Alternatively, it may be set to time t0 to tn indicating the inundation depth distribution group.
(Maximum predictive value during the inundation reduction period)

In the same way, the maximum value of the predictive value in the inundation reduction period is calculated. First, as in FIG. 10, on the horizontal axis, the inundation depth distribution group A0-n when the rainfall intensity j [mm / hr] continues for k hours in the database by rainfall intensity is taken in the order of the inundation depth distribution number, and the vertical axis is appropriate. Create a predictive value graph with the rate. The predictive value graph as shown in FIG. 13 thus obtained is created for each rainfall amount.

Next, from the predictive value graph, the vertical axis of the database by rainfall intensity is converted from the inundation amount to the inundation depth distribution number (any of the group A) of the inundation depth distribution map. Since it is assumed that inland inundation will inundate from the same location in the same order, the ● part in the database by rainfall intensity in Fig. 6 is one of Group A (inundation depth from the start of inundation to the assumed maximum scale). Equivalent to. In this way, the pre-calculated graph of FIG. 14 is obtained.

In this way, the correlation between the results for each rainfall intensity and the results for calculating the basic unit can be obtained. That is, a pre-calculation graph is created in advance and saved in the database. Then, during the actual operation of real-time inundation inundation prediction, the inundation area is predicted by referring to these stored databases. Here, inundation is predicted based on the observed rainfall (past rainfall) and the predicted rainfall (future rainfall) announced by the Japan Meteorological Agency.
(Procedure for actual operation when the water level gauge is not used)

Next, a procedure for estimating the current inundation area and predicting the future inundation area in the actual operation of the inundation inundation prediction according to the first embodiment will be described with reference to the flowchart of FIG. In the real-time inundation inundation prediction according to the first embodiment, it is assumed that the water level gauge is not installed in the specific area as described above.
(S1: Acquisition of rainfall waveform)

First, in step S1, a rainfall waveform is acquired as input data. Specifically, the rainfall data is acquired from the rainfall data acquisition unit 10. Here, real-time observed rainfall and predicted rainfall obtained from the Japan Meteorological Agency are used. The rainfall waveform from the beginning of the rainfall can be obtained by combining the past observed rainfall and the predicted rainfall in the future. An example of such a rainfall waveform is shown in the graph of FIG. Based on this rainfall waveform, real-time inundation inundation prediction is started from the time when the rainfall starts. Here, we focus on the inundation depth distribution at an arbitrary time tw, which is in the future from the current time. Aold, which is the initial value of the inundation depth distribution map, is an inundation depth distribution map at the beginning of descending, and is set to no inundation, that is, the origin (A0).
(S2: Judgment of inundation increase period or inundation decrease period)

Next, in step S2, it is determined whether the future rainfall is in the inundation increase period or the inundation decrease period. Here, the rainfall intensity J [mm / hr] between the time tw and the time tw + 1 is acquired from the Japan Meteorological Agency with reference to an arbitrary time tw. Then, when J> dr, it is determined that the inundation increase period and the process proceeds to step S3-1. On the other hand, when J <dr, it is determined that the inundation reduction period is reached, and the process proceeds to step S3-2.

In step S3-1, inundation analysis is performed during the inundation increase period, and the future inundation depth distribution is acquired. Here, Anew is estimated as the inundation depth distribution at the future time tw + 1 based on the pre-calculation graph. For example, as a pre-calculation graph, it is assumed that the result of the inundation analysis in the inundation increase period in which the rainfall waveform in which the rainfall intensity of J [mm / hr] continues to fall is used as an external force is obtained as the waveform shown in FIG. In this graph, the inundation depth distribution map A at time tw is the position Aw shown on the vertical axis. Let this be Aold, and the elapsed time (time tw) at this time is the position on the corresponding horizontal axis on the pre-calculation graph. In the first loop, as described in step S1, Aold, which is the initial value of the inundation depth distribution map, becomes the origin (A0).

Here, assuming that Δw elapses from the time tw + 1, the position Aw + 1 on the vertical axis corresponding to the position moved to the left by Δw on the horizontal axis is the inundation depth distribution map Aw at the next time tw + 1. It is estimated that it corresponds to +1. The future inundation depth distribution map Aw + 1 predicted in this way is set as Anew, and the process proceeds to step S4.

On the other hand, if it is determined in the inundation reduction period in step S2, the process proceeds to step S3-2, the inundation analysis in the inundation reduction period is performed, and the future inundation depth distribution is obtained in the same manner. Here, too, Anew is estimated as the inundation depth distribution at the future time tw + 1 based on the pre-calculation graph. For example, as a pre-calculation graph, it is assumed that the result of the inundation analysis in the inundation reduction period using the rainfall waveform in which the rainfall intensity of J [mm / hr] continues to fall is obtained as the waveform as shown in FIG. In this graph, the inundation depth distribution map A at time tw is the position Aw shown on the vertical axis. Let this be Aold, and the elapsed time (time tw) at this time is the position on the corresponding horizontal axis on the pre-calculation graph. As described above, where Aold is A0 in the first loop, there is no need to predict inundation inundation because no further inundation reduction period can occur in A0 without inundation.

Here, assuming that Δw elapses from the time tw + 1, the position Aw + 1 on the vertical axis corresponding to the position moved to the left by Δw on the horizontal axis is the inundation depth distribution map Aw at the next time tw + 1. It is estimated that it corresponds to +1. The future inundation depth distribution map Aw + 1 predicted in this way is set as Anew, and the process proceeds to step S4.

Then, in step S4, the inundation depth distribution map Aw + 1 at the time tw + 1 thus obtained is displayed on the display unit 40 as a future inundation depth distribution map Anew.

Further, in step S5, the time shifts to the next time. Here, the time tw is further incremented to w + 1, and the inundation depth distribution map Anew is set to Aold. Then, the process returns to step S2 and the above steps are repeated.

In this way, it is possible to predict the future inundation depth distribution map in real time and sequentially display it on the display unit 40. In particular, by creating and recording a pre-calculation graph in advance in the flood analysis phase and referring to it during actual operation, it is possible to instantly recall the flood analysis results of complex terrain, resulting in a small computational load. However, it is possible to predict inundation and inundation in real time.
[Embodiment 2]
(When a water level gauge is installed)

The above has described an example in which a water level gauge is not installed in a specific area. However, the present invention is not limited to the configuration in which the water level gauge is not installed, and a water level gauge may be installed. By installing a water level gauge, it is possible to realize more accurate real-time inundation inundation prediction by referring to the water level gauge inundation depth database that can specify the inundation start time of the basic unit calculation at the water level observation point. Here, a real-time inundation inundation prediction when a water level gauge is used as the second embodiment will be described.

First, as the flood analysis phase, basic unit calculation and database creation by rainfall intensity are performed. In addition, as the statistical analysis phase, the correlation between the rainfall intensity and the basic unit calculation is obtained based on the basic unit calculation and the database for each rainfall intensity obtained in the flood analysis phase. Specifically, a predictive value graph is created, and a pre-calculation graph is created based on this predictive value graph by converting the vertical axis of the database by rainfall intensity into the inundation depth distribution number of the inundation depth distribution map. This procedure is the same as that of the first embodiment described above, and detailed description thereof will be omitted.
(Process of specifying the inundation start time at the water level observation point)

In addition, in the second embodiment, the inundation start time at the water level observation point is specified as the inundation analysis phase. Here, when the predicted area (or the mesh of the corresponding area) where the inundation is predicted is flooded, the time from the start of rain until the water level reaches the specified water level is defined as the arrival time. Then, the arrival time at the water level observation point is calculated in advance by flood analysis. In addition, in the basic unit calculation, the time when the water level observation point is inundated (the number in the inundation depth distribution group) is associated with the water level gauge. For example, as shown in FIG. 19, when the water level gauge Cx is installed at the water level observation point of the mesh that divides the prediction area where the inundation prediction is performed, the inundation depth distribution map is higher than the specified water level set by this water level gauge. Consider the case where the water level of Ap-1 is low and the water level of the inundation depth distribution map Ap is high. In this case, inundation is not detected in the inundation depth distribution map Ap-1 until the specified water level is reached. Then, in the next inundation depth distribution map Ap, the detected water level detected by the inundation meter has reached the specified water level. Therefore, the water level gauge Cx and the inundation depth distribution map Ap at the pth minute are linked. The table in which the correspondence is recorded in this way is called a water level gauge inundation depth database, and is stored in the water level gauge inundation depth database storage unit 25.
(Procedure for actual operation when using a water level gauge)

Next, in the actual operation of the inundation inundation prediction according to the second embodiment, the procedure for estimating the current inundation area and predicting the future inundation area is performed based on the flowchart of FIG. 15 as in the first embodiment. explain. In the second embodiment, since the water level gauge is installed in the specific area where the inundation analysis is performed as described above, the inundation of the inland waters is predicted based on the water level observation observed by the water level gauge.
(S1: Acquisition of rainfall waveform)

First, in step S1, a rainfall waveform is acquired as input data. Here, as in the first embodiment described above, the rainfall waveform from the beginning of rainfall is obtained by combining the real-time observed rainfall and the predicted rainfall obtained from the Japan Meteorological Agency. An example of such a rainfall waveform is shown in the graph of FIG. Here, unlike the first embodiment, the real-time inundation inundation prediction is not started from the time when the water starts to descend, but is started from the water level gauge reaction time tp, which is the time when the water level gauge Cx reacts. Further, Aold, which is the initial value of the inundation depth distribution map, is not the inundation depth distribution map A0 at the beginning of descending, but the inundation depth distribution map Ap at the water level gauge reaction time tp.
(S2: Judgment of inundation increase period or inundation decrease period)

Next, in step S2, it is determined whether the future rainfall is in the inundation increase period or the inundation decrease period. Here, too, the rainfall intensity J [mm / hr] between the time tw and the time tw + 1 is obtained from the Japan Meteorological Agency with reference to an arbitrary time tw. Then, when J> dr, it is determined that the inundation increase period and the process proceeds to step S3-1. On the other hand, when J <dr, it is determined that the inundation reduction period is reached, and the process proceeds to step S3-2.

In step S3-1, inundation analysis is performed during the inundation increase period, and the future inundation depth distribution is acquired. Here, Anew is estimated as the inundation depth distribution at the future time tw + 1 based on the pre-calculation graph. For example, as a pre-calculation graph, it is assumed that the result of the inundation analysis in the inundation increase period in which the rainfall waveform in which the rainfall intensity of J [mm / hr] continues to fall is used as an external force is obtained as the waveform shown in FIG. In this graph, the inundation depth distribution map A at time tw is the position Aw shown on the vertical axis. Let this be Aold, and the elapsed time (time tw) at this time is the position on the corresponding horizontal axis on the pre-calculation graph. In the first embodiment, Aold was A0 in the first loop, but as described above, in the second embodiment, Aold is Ap in the first loop. After that, it is the same as the first embodiment.

Even when a water level gauge is used, the current inundation area is predicted by the time change (hyetograph) of the observed rainfall before the observed water level by the water level gauge is obtained, that is, before the water level gauge reacts. That is, the process is the same as that of the first embodiment described above. When the water level gauge is operating, the current inundation area estimated from the operating status of the water level gauge is estimated. After that, the future inundation area is predicted using the predicted rainfall in the future. Furthermore, if one water level gauge reacts and then another water level gauge installed at another water level gauge position reacts newly, the inland water flooding will occur based on the water level observation observed by this water level gauge. Predict inundation. In this way, by updating with the latest reaction results of the water level gauge, it is possible to predict the inundation inundation more accurately, reflecting the latest actual inundation situation. As a result, it is possible to obtain inland inundation and inundation prediction results with little error.
(Details of flood analysis of Embodiment 1)

Here, the details of the inundation analysis in the real-time inundation inundation inundation prediction according to the first embodiment in which the water level gauge is not installed will be described. Now, it is assumed that the graph of FIG. 21 is obtained as the rainfall waveform. This example shows the situation at 7 o'clock. Therefore, the rainfall from 4 o'clock to 7 o'clock is the past observed rainfall, which is indicated by a crosshatch in the figure. In addition, 7:00 to 9:00 is the predicted rainfall in the future, which is indicated by a one-sided hatch in the figure. Furthermore, this example shows that it started to rain at 4 o'clock.

Here, a state in which the inundation depth distribution of the basic unit calculation result is displayed on the display unit 40 by using the pre-calculation graph and the past observed rainfall / future predicted rainfall from the beginning of the rainfall will be described with reference to FIGS. 22A to 22E.

First, at the beginning of the descent at 4 o'clock, the display unit 40 displays the inundation depth distribution at the 0th minute in the basic unit calculation as the inundation depth distribution at 4 o'clock. As shown in FIG. 22A, the start of descending (4 o'clock) corresponds to the 0th minute.

Next, an example in which the display unit 40 displays the inundation depth distribution at 5 o'clock one hour after the start of descending will be described. Here, since the rainfall amount from 4 o'clock to 5 o'clock is 15 mm / hr in FIG. 21, a pre-calculation graph having a rainfall amount of 15 mm / hr is extracted from the pre-calculation graph database as a pre-calculation graph. Then, as shown in FIG. 22A, in the pre-calculation graph of 15 mm / hr, the horizontal axis is slid from the origin for 1 hour from 4 o'clock to 5 o'clock. Here, when it is confirmed how many minutes the inundation area corresponds to the time point of sliding for 1 hour in the basic unit calculation, it can be seen that it corresponds to the 15th minute on the vertical axis. According to this, the display unit 40 displays the inundation depth distribution at the 15th minute in the basic unit calculation. In other words, it can be seen that the inundation depth distribution at 5 o'clock corresponds to the 15th minute in the basic unit calculation.

Further, an example of displaying the inundation depth distribution at 6 o'clock 2 hours after the start of descending on the display unit 40 will be described. First, since the rainfall amount from 5 o'clock to 6 o'clock is 20 mm / hr according to FIG. 21, a pre-calculation graph of the rainfall amount of 20 mm / hr is extracted from the pre-calculation graph database. Then, as shown in FIG. 22B, in the pre-calculation graph of 20 mm / hr, the position corresponding to the inundation depth distribution at 5 o'clock described above, that is, the position of the inundation area at the 15th minute in the basic unit calculation is specified on the vertical axis. , The corresponding elapsed time is specified on the horizontal axis. Since this position corresponds to 5 o'clock, slide it on the horizontal axis for 1 hour from here. At this position, the position of the corresponding vertical axis is the 23rd minute. That is, it can be seen that the inundation depth distribution at 6 o'clock corresponds to the inundation depth distribution at the 23rd minute in the basic unit calculation. Therefore, by displaying the inundation depth distribution at this time on the display unit 40, the inundation depth distribution at 6 o'clock can be visually confirmed.

Furthermore, an example of displaying the inundation depth distribution at 7 o'clock 3 hours after the start of descending on the display unit 40 will be described. First, since the rainfall amount from 6:00 to 7:00 is 25 mm / hr according to FIG. 21, a pre-calculation graph of the rainfall amount of 25 mm / hr is extracted from the pre-calculation graph database. Then, as shown in FIG. 22C, in the pre-calculation graph of 25 mm / hr, the position corresponding to the inundation depth distribution at 6 o'clock described above, that is, the position of the inundation area at the 23rd minute in the basic unit calculation is specified on the vertical axis. , The corresponding elapsed time is specified on the horizontal axis. Since this position corresponds to 6 o'clock, slide it on the horizontal axis for 1 hour from here. At this position, the position of the corresponding vertical axis is the 34th minute. That is, it can be seen that the inundation depth distribution at 7 o'clock corresponds to the inundation depth distribution at the 34th minute in the basic unit calculation. Therefore, by displaying the inundation depth distribution at this time on the display unit 40, the inundation depth distribution at 7 o'clock can be visually confirmed.

Furthermore, an example of displaying the inundation depth distribution at 8 o'clock 4 hours after the start of descending on the display unit 40 will be described. In the above example, an example of displaying the past inundation depth distribution has been described, but thereafter, the future inundation depth distribution, that is, the inundation area prediction will be used. First, since the rainfall amount from 7:00 to 8:00 is 20 mm / hr according to FIG. 21, a pre-calculation graph of the rainfall amount of 20 mm / hr is extracted from the pre-calculation graph database. In the above example, the amount of rainfall increased, but here the amount of rainfall has decreased. As shown in FIG. 22D, in the 20 mm / hr pre-calculation graph, the position corresponding to the inundation depth distribution at 7 o'clock described above, that is, the position of the inundation area at the 34th minute in the basic unit calculation is specified on the vertical axis. The corresponding elapsed time is specified on the horizontal axis. Since this position corresponds to 7 o'clock, if it is slid on the horizontal axis for 1 hour from here, the position of the corresponding vertical axis is the 37th minute. That is, it can be seen that the inundation depth distribution at 8 o'clock corresponds to the inundation depth distribution at the 37th minute in the basic unit calculation. Therefore, by displaying the inundation depth distribution at this time on the display unit 40, the inundation depth distribution at 8 o'clock can be visually confirmed.

Further, an example of displaying the inundation depth distribution at 9 o'clock 5 hours after the start of descending on the display unit 40 will be described. First, since the rainfall amount from 8:00 to 9:00 is 5 mm / hr according to FIG. 21, a pre-calculation graph of the rainfall amount of 5 mm / hr is extracted from the pre-calculation graph database. It can be seen that the amount of rainfall is further reduced. Then, as shown in FIG. 22E, in the pre-calculation graph of 5 mm / hr, the position corresponding to the inundation depth distribution at 8 o'clock described above, that is, the position of the inundation area at the 37th minute in the basic unit calculation is specified on the vertical axis. , The corresponding elapsed time is specified on the horizontal axis. Since this position corresponds to 8 o'clock, slide it on the horizontal axis for 1 hour from here. At this position, the position of the corresponding vertical axis is the 28th minute. That is, it can be seen that the inundation depth distribution at 9 o'clock corresponds to the inundation depth distribution at the 28th minute in the basic unit calculation. Therefore, by displaying the inundation depth distribution at this time on the display unit 40, the inundation depth distribution at 9 o'clock can be visually confirmed.

As described above, it is possible to display not only the estimation of the past inundation depth distribution but also the future inundation area prediction on the display unit 40 in real time. That is, the corresponding inundation depth distribution can be obtained from the inundation depth distribution database simply by specifying which of the inundation depth distributions prepared in advance by calculation without going through complicated calculations such as ground surface analysis. Since it is sufficient to read and display it, it is possible to significantly reduce the calculation cost and display an accurate inundation depth distribution with a light load.
(Details of flood analysis of Embodiment 2)

In addition, the details of the inundation analysis in the real-time inundation inundation inundation prediction according to the second embodiment in which the water level gauge is installed will be described. Here, too, it is assumed that the graphs of the past observed rainfall and the future predicted rainfall having the same waveform as those in FIG. 21 described above are obtained from the Japan Meteorological Agency.

In FIG. 21, assuming that the water level gauge reacts and reaches the specified water level at 7 o'clock, which is the current time, an example of displaying the inundation depth distribution at this time, that is, the current inundation depth distribution will be described. For each water level gauge, the arrival time until the water level reaches the specified water level at the water level observation point where the water level gauge is installed is known in advance. That is, by referring to the water level gauge inundation depth database that specifies the inundation start time at the water level observation point from the water level gauge inundation depth database storage unit 25, the correspondence between the water level gauge Cx and the inundation depth distribution Ap at the pth minute can be determined. Can be identified. For example, it is assumed that the arrival time at the water level observation point of the reacted water level gauge Cx is 42 minutes in the basic unit calculation. In this case, as the inundation depth distribution at 7 o'clock, the inundation depth distribution A42 at the 42nd minute in the basic unit calculation is displayed on the display unit 40.

Next, a method for predicting a future inundation area will be described. First, an example of displaying the inundation depth distribution at 8 o'clock one hour later will be described. First, the rainfall from 7:00 to 8:00 is 20 mm / hr according to the future forecast rainfall in FIG. 21. Therefore, a pre-calculation graph with a rainfall of 20 mm / hr is extracted from the pre-calculation graph database. Then, as shown in FIG. 23A, in the pre-calculation graph of 20 mm / hr, the position corresponding to the inundation depth distribution at 7 o'clock described above, that is, the position of the inundation area at the 42nd minute in the basic unit calculation is specified on the vertical axis. , The corresponding elapsed time is specified on the horizontal axis. Since this position corresponds to 7 o'clock, slide it on the horizontal axis for 1 hour from here. Then, at the position one hour later, the position of the corresponding vertical axis is the 46th minute. That is, the inundation depth distribution at 8 o'clock is expected to correspond to the inundation depth distribution at the 46th minute in the basic unit calculation. Therefore, by reading the inundation depth distribution at this time from the inundation depth distribution database and displaying it on the display unit 40, the inundation depth distribution at 8 o'clock can be visually confirmed.

Similarly, an example of displaying the inundation depth distribution at 9 o'clock after 2 hours will be described. First, since the rainfall from 8:00 to 9:00 is 5 mm / hr according to the future forecast rainfall in FIG. 21, a pre-calculation graph of rainfall of 5 mm / hr is extracted from the pre-calculation graph database. Then, as shown in FIG. 23B, in the pre-calculation graph of 5 mm / hr, the position corresponding to the inundation depth distribution at 8 o'clock described above, that is, the position of the inundation area at the 46th minute in the basic unit calculation is specified on the vertical axis. , The corresponding elapsed time is specified on the horizontal axis. Since this position corresponds to 8 o'clock, slide it on the horizontal axis for 1 hour from here. Then, at the position one hour later, the position of the corresponding vertical axis is the 31st minute. That is, the inundation depth distribution at 9 o'clock is expected to correspond to the inundation depth distribution at the 31st minute in the basic unit calculation. Therefore, by reading the inundation depth distribution at this time from the inundation depth distribution database and displaying it on the display unit 40, the inundation depth distribution at 9 o'clock can be visually confirmed.
(Update of future inundation area forecast)

Further, the rainfall data at 8 o'clock one hour after 7 o'clock in FIG. 21 is shown in FIG. 24. Here, the past observed rainfall observed before 7 o'clock remains the same, and the past 1 hour's worth of rainfall, that is, the past observed rainfall from 7 o'clock to 8 o'clock and the future predicted rainfall from 8 o'clock to 9 o'clock are updated from Fig. 21. Has been done. Here, an example of predicting the past inundation depth distribution and the future inundation area as of 8 o'clock will be described with reference to FIGS. 25A to 25B.

First, an example of displaying the past inundation depth distribution at 7 o'clock one hour ago will be described. Here, as in the above method, since the arrival time at the water level observation point where the water level gauge that reacted at 7 o'clock one hour ago was installed is 42 minutes in the basic unit calculation, the inundation depth distribution at 7 o'clock is used. The inundation depth distribution A42 at the 42nd minute in the basic unit calculation is displayed on the display unit 40.

Next, a method of displaying the current inundation depth distribution at 8 o'clock will be described. First, the rainfall from 7:00 to 8:00 is 25 mm / hr according to the past observed rainfall in FIG. 24. Therefore, a pre-calculation graph with a rainfall of 25 mm / hr is extracted from the pre-calculation graph database. Then, as shown in FIG. 25A, in the pre-calculation graph of 25 mm / hr, the position corresponding to the inundation depth distribution at 7 o'clock described above, that is, the position of the inundation area at the 42nd minute in the basic unit calculation is specified on the vertical axis. , The corresponding elapsed time is specified on the horizontal axis. Since this position corresponds to 7 o'clock, slide it on the horizontal axis for 1 hour from here. Then, at the position one hour later, the position of the corresponding vertical axis is the 49th minute. That is, the inundation depth distribution at 8 o'clock is estimated to correspond to the inundation depth distribution at the 49th minute in the basic unit calculation. Therefore, by reading the inundation depth distribution at this time from the inundation depth distribution database and displaying it on the display unit 40, the inundation depth distribution at 8 o'clock can be visually confirmed. Compared with FIG. 23A described above, it can be seen that since there was more rainfall than expected at 7 o'clock, the inundation depth distribution was updated to a higher water level. As described above, according to the present embodiment, it can be understood that the display of the current inundation depth distribution is corrected to more accurate information reflecting the actual rainfall.

Next, an example of displaying the inundation depth distribution at 9 o'clock one hour after 8 o'clock will be described. First, since the rainfall from 8:00 to 9:00 is 8 mm / hr according to the future forecast rainfall in FIG. 24, a pre-calculation graph of the rainfall of 8 mm / hr is extracted from the pre-calculation graph database. Then, as shown in FIG. 25B, in the pre-calculation graph of 8 mm / hr, the position corresponding to the inundation depth distribution at 8 o'clock described above, that is, the position of the inundation area at the 49th minute in the basic unit calculation is specified on the vertical axis. , The corresponding elapsed time is specified on the horizontal axis. Since this position corresponds to 8 o'clock, slide it on the horizontal axis for 1 hour from here. Then, at the position one hour later, the position of the corresponding vertical axis is the 35th minute. That is, the inundation depth distribution at 9 o'clock is expected to correspond to the inundation depth distribution at the 35th minute in the basic unit calculation. Therefore, by reading the inundation depth distribution at this time from the inundation depth distribution database and displaying it on the display unit 40, the inundation depth distribution at 9 o'clock can be visually confirmed. As described above, according to the present embodiment, it can be understood that the prediction of the future inundation depth distribution is also corrected to more accurate information by reflecting the newer future prediction rainfall.

In this way, until another water level gauge reacts, the inundation depth distribution can be performed with the most recently reacted water level gauge as the initial condition. In addition, when a new water level gauge reacts, it is updated to specify the inundation depth distribution based on this water level gauge. As a result, it is possible to update the prediction of the inundation depth distribution based on the most recent water level, and it is possible to improve the accuracy.

When a plurality of water level gauges react at the same time, the water level of the water level gauge having a large subscript value of group A or a large subscript value of ti is adopted. In this way, by adopting a more dangerous inundation depth distribution map, it is possible to predict a higher water level and improve safety.
[Embodiments 3 and 4]

In the above example, the inundation depth distribution was calculated using the pound model, which is a flood analysis that does not consider the water surface gradient. However, the present invention does not limit the method for determining the inundation depth distribution to the pound model, and other known methods can also be applied. For example, the inundation depth distribution with a water surface gradient obtained from the inundation analysis by the plane two-dimensional indefinite flow calculation may be used. In this specification, in the area subject to inland inundation inundation prediction, the prediction is made on the premise that inundation will occur in the same order from the same location. However, when using inundation analysis that does not consider the water surface gradient, It is assumed that the water will be flooded in order from the lowest altitude. On the other hand, when applying a flood analysis that can consider the water surface gradient, the premise that inundation occurs in the same order from the same location means that inundation occurs in the order of the result from the inundation location in the calculation result of the basic unit calculation. Become. As described above, in the present specification, the meaning of "flooding from the same place in the same order" means "flooding from the flooded place in the calculation result of the basic unit calculation in the order of the result".

Here, instead of the pound model, an example of performing inundation analysis using an inundation depth distribution with a water surface gradient, which is obtained from inundation analysis by plane two-dimensional indefinite flow calculation, is an example of inland inundation inundation according to Embodiments 3 and 4. The prediction method will be described below. Here, the case where the water level gauge is present will be described as the third embodiment, and the case where the water level gauge is not provided will be described as the fourth embodiment. In the method for predicting inundation and inundation according to the third and fourth embodiments, it is assumed that rainfall is applied to the region shown in FIG. 26 and that there is a displacement equivalent ^{to 0.04 m 3 / sec from this region.}
(1: Basic unit calculation)

The rainfall waveform given as an external force (input) in the basic unit calculation is shown in the graph of FIG. 27. Further, the amount of inundation, which is the result of the basic unit calculation (output) when the rainfall waveform of FIG. 27 is given, is shown in FIG. 28, and the water level is shown in FIG. 29, respectively.
(2: Creation of database by rainfall intensity)
1) Water level increase period

FIG. 30 shows the time change of the water level during the water level increase period. In this figure, the time change of the water level when the rainfall waveform is used as an external force when each rainfall continues to fall without changing the time is shown every 5 mm / hr from 15 mm / hr to 100 mm / hr.
2) Water level decrease period

FIG. 31 shows the time change of the water level during the water level decrease period. FIG. 31 shows the time change of the water level when the rainfall waveform in which the rainfall continues to fall without changing the time is used as an external force every 5 mm / hr from 0 mm / hr to 10 mm / hr.
(3: Accuracy rate)

Next, in order to obtain the accuracy rate, the vertical axis of FIGS. 30 and 31 is converted into group A (inundation depth distribution group). In the third and fourth embodiments, since it is assumed that the water is flooded from the lowland in a state where there is no water surface gradient, the relationship between the inundation amount and the water level is represented by Equation 1. Therefore, the inundation amount is converted into a water level using Equation 2, and the inundation depth distribution in which the lowland is inundated below the water level is output as the output of the real-time inundation prediction system.
[Number 2]
V = 1/3 (H-1) ³

The vertical axis of FIGS. 30 and 31 is group A (inundation depth distribution group), which are shown in FIGS. 32 and 33, respectively. In the third and fourth embodiments, the basic unit calculation was output every 10 minutes. Therefore, the period from Aw to Aw + 1 corresponds to 10 min in the basic unit calculation. In the basic unit calculation, after A120, the water level changed from increasing to decreasing. Therefore, since A120 is considered to correspond to the assumed maximum inundation depth distribution, the inundation depth distribution with a larger inundation scale than A120 was set to be equivalent to A120.
(4: Inundation start time of basic unit calculation at water level observation point)

Here, it is assumed that the water level gauge reacts when the water level reaches 13.0 [m]. It has been found in the basic unit calculation that Ap at that time corresponds to A10.
(5: When not using the water level gauge)

First, consider the real-time operation when the water level gauge is not used. Here, the time change of the inundation depth distribution when the rainfall waveform shown in FIG. 34 is applied as an external force will be examined. Data provided by the Japan Meteorological Agency can be used for such rainfall waveforms.

The time change of the output inundation depth distribution group (Group A) is shown in FIG. 35. When the water level gauge described later is used, the inundation inundation prediction is started from the time when the water level gauge reacts, whereas when the water level gauge is not installed, it is started from the rainfall start time. From FIG. 35, it can be seen that the calculated value by the pound model and the tendency of the water level increase / decrease in the calculation by the inland inundation inundation prediction method according to the third embodiment are almost the same. The vertical axis of the graph shown in FIG. 35 is the numerical value of the subscript of group A (inundation depth distribution group). It can be seen that the water level rises to the water level corresponding to A12 at the maximum inundation condition by the pound model, whereas the water level rises to the water level corresponding to A13 in the inland inundation inundation prediction method according to the third embodiment. ..
(6: When using a water level gauge)

Next, FIG. 36 shows the time change of the inundation depth distribution group (group A) output when the water level gauge is used. Here, the prediction of inundation and inundation starts from the time when the water level gauge reacts. From FIG. 36, it can be seen that the calculated value by the pound model and the tendency of the water level increase / decrease in the calculation according to the fourth embodiment are substantially the same. Both the calculated value by the pound model and the calculation according to the fourth embodiment show that the water level rises to the water level corresponding to A12 when the inundation condition is the maximum. It is considered that the maximum condition of the inundation scale was accurately determined by considering the observed water level.

The present invention is not limited to the method using the predictive value, and it is possible to apply another method other than the predictive value so that the vertical axis in the database for each rainfall intensity is group A (inundation depth distribution group). can. here,
(I) Inundation depth distribution map (II) Ai when rainfall of rainfall intensity J [mm / hr] continues for k hours (part indicated by ● in the figure) in the database for each rainfall intensity shown in FIGS. 5 and 6. Inundation depth distribution group A0 to An, which is a set of inundation depth distribution maps in
Then, a known method for evaluating the degree of similarity, consistency, correlation, compatibility, etc. in the inundation depth distribution maps of (I) and (II) can be appropriately used.

As an alternative method to the accuracy rate, for example, as shown in the graph of FIG. 38, the horizontal axis is the inundation depth [m] in the inundation depth distribution of (I), and the vertical axis is the inundation depth distribution map of (II). Taking the inundation depth [m], the inundation depths of the inundation depth distribution maps of (I) and (II) at the same location are plotted. The total value Σd of each plot at the distance d between each plot obtained in this way and the straight line having an inclination of 45 ° is calculated, and Ai, which is the minimum of Σd, is defined as Am. Thereby, the vertical axis in the database by rainfall intensity can be converted into the inundation depth distribution groups A0 to An.

In addition, as another method instead of the accuracy rate, as shown in the graph of FIG. 39, the difference in inundation depth in the inundation depth distribution maps of (I) and (II) at the same location is examined. Here, the difference [m] in the inundation depths of the inundation depth distribution maps (I) and (II) at the same location is examined, the number of meshes is counted, and a graph in which the distribution is examined is created. Then, by calculating the average value μ and setting Ai whose average value μ is closest to 0 as Am, the vertical axis in the rainfall intensity database can be converted into the inundation depth distribution groups A0 to An.

According to such a known method of showing the degree of correlation between (I) and (II), the vertical axis of the database for each rainfall intensity can be converted into group A.

Using the real-time inland inundation inundation prediction system of the present invention, real-time inland inundation inundation prediction device, real-time inland waters inundation inundation prediction method, real-time inland waters inundation inundation prediction program, computer-readable recording medium, and stored equipment. , It becomes possible to predict inundation and inundation. As a result, when a super-large typhoon or guerrilla rainstorm occurs, the occurrence of inland water inundation in urban areas can be accurately predicted, which can be useful for creating evacuation guide routes.

1000, 1000'... Real-time inundation inundation prediction system 100 ... Real-time inland inundation inundation prediction device 10 ... Input unit 20 ... Data storage unit 21 ... Inundation depth distribution group storage unit 22 ... Pre-calculation graph storage unit 23 ... By rainfall intensity Database storage unit 24 ... Accuracy rate graph storage unit 25 ... Water level gauge Inundation depth database storage unit 30 ... Calculation unit 31 ... Inundation depth prediction unit 32 ... Database access unit 33 ... Inundation depth distribution group creation unit 34 ... Pre-calculation graph creation unit 40 ... Display unit 41 ... Map display area 50 ... Operation unit CL ... Client device WG ... Water level gauge RT ... Repeater

Claims (13)

Real-time inland waters that predict inland inundation caused by rainfall in real time under the assumption that uniform rainfall will occur in the predicted area and that inundation will occur in a certain order at each position included in the predicted area. Inundation prediction system,
As a set of inundation depth distribution map Ai showing the inundation status of the predicted area at a specific time ti, it is assumed from the inundation depth distribution map A0 in which inundation does not occur in the predicted area at the inundation start time t0 at predetermined time intervals. Inundation depth distribution map at time tn, which is the largest scale Inundation depth distribution group preservation section that preserves the inundation depth distribution group up to An, and the inundation depth distribution group preservation section.
Among the inundation depth distribution groups stored in the inundation depth distribution group preservation section, the rainfall intensity of j [mm / hr] falls between k [hr] under the initial condition of the inundation depth distribution map at a predetermined time. In this case, the correlation between the rainfall intensity and the inundation depth distribution map created from the rainfall intensity-specific database showing the number of the inundation depth distribution in the inundation depth distribution group A (inundation depth distribution) is shown. A pre-calculation graph saver that saves the pre-calculation graph shown,
The rainfall data acquisition unit that acquires the observed rainfall and the predicted rainfall,
Based on the observed rainfall acquired by the rainfall data acquisition unit and the predicted rainfall, a future inundation depth distribution map can be obtained from a plurality of inundation depth distribution maps stored in the inundation depth distribution group preservation unit. , The inundation depth prediction unit selected based on the correspondence defined by the pre-calculation graph stored in the pre-calculation graph storage unit,
A display unit that displays an inundation depth distribution map selected by the inundation depth prediction unit, and a display unit that displays the inundation depth distribution map.
Real-time inundation inundation prediction system equipped with. The real-time inundation inundation prediction system according to claim 1.
A real-time inundation inundation inundation prediction system in which the inundation depth prediction unit uses an inundation depth distribution map at a predetermined time as an initial condition as an inundation depth distribution map A0 at a time t0 when there is no inundation. The real-time inundation inundation prediction system according to claim 1, further
A water level gauge for detecting the water level, which is installed at each of the multiple water level observation points set in the forecast area,
A water level gauge that associates a plurality of inundation depth distribution maps A0 to An stored in the inundation depth distribution group storage unit with a water level gauge reaction time tp at which each water level gauge installed at the plurality of water level observation points Cx reacts. Water level gauge that stores the inundation depth database Inundation depth database storage unit,
Is equipped with
The inundation depth prediction unit makes the inundation depth distribution map at a predetermined time as an initial condition a corresponding inundation depth distribution map Ap at the water level gauge reaction time tp in which any of the plurality of water level gauges reacts. , A real-time inundation inundation prediction system specified from the water level gauge inundation depth database. The real-time inundation inundation prediction system according to claim 3.
The inundation depth prediction unit is configured to select the inundation depth distribution map by adopting the reaction of the water level gauge with the higher detected water level when a plurality of water level gauges react at the same time. Inundation inundation prediction system. The real-time inundation inundation prediction system according to any one of claims 1 to 4.
A real-time inundation inundation prediction system in which the database by rainfall intensity is a graph in which the horizontal axis is the elapsed time and the vertical axis is the amount of inundation. The real-time inundation inundation prediction system according to claim 5.
The pre-calculation graph storage unit
Inundation increase period when the external rainfall force is larger than the drainage capacity in the predicted area,
A real-time inundation inundation prediction system including a rainfall intensity database storage unit that stores the rainfall intensity-specific database for each of the inundation reduction periods in which the external rainfall force is smaller than the drainage capacity in the predicted area. The real-time inundation inundation prediction system according to claim 5 or 6.
The pre-calculation graph stored in the pre-calculation graph storage unit is a graph in which the vertical axis of the database by rainfall intensity is converted from the inundation amount to the inundation depth distribution groups A0 to An. Real-time inundation inundation prediction system. The real-time inundation inundation prediction system according to any one of claims 1 to 7.
The pre-calculation graph storage unit further
The horizontal axis is the inundation depth distribution group A0-n when the rainfall intensity j [mm / hr] continues for k hours in the database by rainfall intensity, and the inundation depth distribution group A0-n is taken in the order of the inundation depth distribution number, and the vertical axis is the accuracy rate graph. Includes a predictive value graph saver to save
The pre-calculation graph stored in the pre-calculation graph storage unit is obtained from the predictive value graph stored in the predictive value graph storage unit, the vertical axis of the database by rainfall intensity, the inundation amount, and the inundation. A real-time inundation inundation prediction system that is converted to the inundation depth distribution number of the depth distribution map. The real-time inundation inundation prediction system according to any one of claims 1 to 8, further comprising.
The inundation depth distribution group creation unit that creates the inundation depth distribution group, and
The pre-calculation graph creation unit that creates the pre-calculation graph,
Real-time inundation inundation prediction system equipped with. Real-time inland waters that predict inland inundation caused by rainfall in real time under the assumption that uniform rainfall will occur in the predicted area and that inundation will occur in a certain order at each position included in the predicted area. It is a flood inundation prediction device,
As a set of inundation depth distribution map Ai showing the inundation status of the predicted area at a specific time ti, it is assumed from the inundation depth distribution map A0 in which inundation does not occur in the predicted area at the inundation start time t0 at predetermined time intervals. Inundation depth distribution map at time tn, which is the largest scale Inundation depth distribution group preservation section that preserves the inundation depth distribution group up to An,
Among the inundation depth distribution groups stored in the inundation depth distribution group preservation section, the rainfall intensity of j [mm / hr] falls between k [hr] under the initial condition of the inundation depth distribution map at a predetermined time. In this case, the correlation between the rainfall intensity and the inundation depth distribution map created from the rainfall intensity-specific database showing the number of the inundation depth distribution in the inundation depth distribution group A (inundation depth distribution) is shown. A pre-calculation graph saver that saves the pre-calculation graph shown,
With a database access section that can access
The rainfall data acquisition unit that acquires the observed rainfall and the predicted rainfall,
Based on the observed rainfall acquired by the rainfall data acquisition unit and the predicted rainfall, a future inundation depth distribution map can be obtained from a plurality of inundation depth distribution maps stored in the inundation depth distribution group preservation unit. , An inundation depth prediction unit that can be selected and output based on the correspondence specified in the pre-calculation graph stored in the pre-calculation graph storage unit.
Real-time inundation inundation prediction device equipped with. Real-time inland waters that predict inland inundation caused by rainfall in real time under the assumption that uniform rainfall will occur in the predicted area and that inundation will occur in a certain order at each position included in the predicted area. It is a method of predicting inundation and inundation.
As a set of inundation depth distribution map Ai showing the inundation status of the predicted area at a specific time ti, it is assumed from the inundation depth distribution map A0 in which inundation does not occur in the predicted area at the inundation start time t0 at predetermined time intervals. The process of creating an inundation depth distribution group up to the inundation depth distribution map An at the time tn, which is the maximum scale, and storing it in the inundation depth distribution group storage section.
Among the inundation depth distribution groups stored in the inundation depth distribution group preservation section, the rainfall intensity of j [mm / hr] falls between k [hr] under the initial condition of the inundation depth distribution map at a predetermined time. In this case, the correlation between the rainfall intensity and the inundation depth distribution map created from the rainfall intensity-specific database showing the number of the inundation depth distribution in the inundation depth distribution group A (inundation depth distribution) is shown. The process of creating the pre-calculation graph to be shown and saving it in the pre-calculation graph storage unit,
The process of acquiring the observed rainfall and the predicted rainfall by the rainfall data acquisition unit,
Based on the observed rainfall acquired by the rainfall data acquisition unit and the predicted rainfall, one of the plurality of inundation depth distribution maps stored in the inundation depth distribution group preservation unit can be used as the future inundation depth. As a distribution map, a process of selecting by the inundation depth prediction unit based on the correspondence defined by the pre-calculation graph stored in the pre-calculation graph storage unit, and
A step of displaying the inundation depth distribution map selected by the inundation depth prediction unit on the display unit, and
Real-time inundation inundation prediction method including. Real-time inland waters that predict inland inundation caused by rainfall in real time under the assumption that uniform rainfall will occur in the predicted area and that inundation will occur in a certain order at each position included in the predicted area. Inundation prediction program,
As a set of inundation depth distribution map Ai showing the inundation status of the predicted area at a specific time ti, it is assumed from the inundation depth distribution map A0 in which inundation does not occur in the predicted area at the inundation start time t0 at predetermined time intervals. A function to create an inundation depth distribution group up to the inundation depth distribution map An at the time tn, which is the maximum scale, and save it in the inundation depth distribution group storage section.
Among the inundation depth distribution groups stored in the inundation depth distribution group preservation section, the rainfall intensity of j [mm / hr] falls between k [hr] under the initial condition of the inundation depth distribution map at a predetermined time. In this case, the correlation between the rainfall intensity and the inundation depth distribution map created from the rainfall intensity-specific database showing the number of the inundation depth distribution in the inundation depth distribution group A (inundation depth distribution) is shown. A function to create the pre-calculation graph to be shown and save it in the pre-calculation graph storage unit,
The function to acquire the observed rainfall and the predicted rainfall by the rainfall data acquisition unit,
Based on the observed rainfall acquired by the rainfall data acquisition unit and the predicted rainfall, one of the plurality of inundation depth distribution maps stored in the inundation depth distribution group preservation unit can be used as the future inundation depth. As a distribution map, a function to be selected by the inundation depth prediction unit based on the correspondence defined by the pre-calculation graph stored in the pre-calculation graph storage unit, and
A function to display the inundation depth distribution map selected by the inundation depth prediction unit on the display unit, and
A real-time inundation inundation prediction program to realize the above on a computer. A computer-readable recording medium or device that records the program of claim 12.

Images