Method for optimizing arrangement of water-level gauge, method for predicting water level in manhole, and system for predicting water level in manhole

Abstract

To predict the water level of a manhole more effectively.SOLUTION: The method for optimizing the arrangement of a water-level gauge includes: a data acquisition step of acquiring water level data acquired by each water-level gauge of a plurality of manholes provided in a predetermined area; a classification step of performing clustering on each piece of water-level data and classifying the manholes into at least one cluster; and a determination step of determining at least one representative manhole according to each cluster from a plurality of manholes included in at least one cluster.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a method for optimizing the arrangement of water level gauges, a method for predicting the water level in a manhole, and a method for predicting the water level in a manhole. This relates to a water level prediction system. [Background technology]

[0002] In recent years, the amount of rainfall from torrential rains and thunderstorms has increased, causing water to overflow from manholes and resulting in flooding. There have been a series of incidents. Therefore, in areas where flooding is expected, It is necessary to measure water level fluctuations (especially rises) in real time and predict flooding. For example, in Patent Document 1, information on the water level in the sewer main, the water level at the pumping station, the amount of rainfall, etc. is used. The paper discloses a technology for predicting water levels and inflows at pumping stations. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2002-285634 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, the technology described in Patent Document 1 is only a prediction of the inflow amount to a pumping station. This is based on the assumption that information on the water level inside the water pipes is used. Therefore, it is not possible to predict overflow onto the ground. In addition, in the prediction of overflow onto the ground, the space connecting the sewer pipe to the ground is It is necessary to monitor the water level in the manholes. However, since there are many manholes, In order to monitor the water level in all manholes and predict the fluctuation of the water level, it is necessary to install many water level gauges. However, this is difficult to achieve.

[0005] Therefore, the present invention has been made in consideration of the above problems, and the object of the present invention is to is a new and improved manhole water level gauge that can more efficiently predict water levels. Method for optimizing manhole layout, method for predicting manhole water level, and system for predicting manhole water level The purpose of this initiative is to provide [Means for solving the problem]

[0006] In order to solve the above problems, according to one aspect of the present invention, there is provided a method for optimizing the arrangement of water level gauges. The water level is obtained by each of the water level gauges installed in a plurality of manholes installed in a specified area. A data acquisition step of acquiring each of the water level data obtained by the data acquisition step; By performing clustering, the plurality of manholes are classified into one or more clusters. and selecting one or more representative manholes from the plurality of manholes included in the one or more clusters. A method for optimizing the arrangement of water level gauges is provided, which includes a step of determining a hole for each cluster. It will be offered.

[0007] In order to solve the above problems, according to another aspect of the present invention, a method for predicting the water level in a manhole is provided. A water level measurement method comprising: obtaining past time-series water level data of a manhole provided in a specified area; A past data acquisition step of acquiring predicted rainfall data relating to the predetermined area; The past time series water level data and the predicted rainfall data of the manhole are input to a predetermined engine. By inputting the information into the input device, the predicted water level of the manhole at a future time point is output. A method for predicting a water level in a manhole is provided, the method comprising the steps of:

[0008] In order to solve the above problems, according to another aspect of the present invention, a method for predicting the water level in a manhole is provided. A water level measurement system that collects water level data from a manhole in a target area and a time series of water levels. a past data acquisition unit that acquires predicted rainfall data relating to the predetermined area; The historical water level data of the manhole and the forecast rainfall data are input to a specified engine. By inputting, the predicted water level of the manhole at a future time point is output. A manhole water level prediction system is provided, the system including: Effect of the Invention

[0009] As described above, according to the present invention, it is possible to more efficiently predict the water level in a manhole. It is possible. [Brief description of the drawings]

[0010] [Figure 1] FIG. 1 is an explanatory diagram for explaining an overview of an information processing system according to an embodiment of the present invention. [Diagram 2] 1 is a block diagram showing an example of a functional configuration of a server 10 according to a first embodiment of the present invention. [Diagram 3] FIG. 2 is a diagram showing an example of clustering processing of manholes 4 by a classification unit 102 according to the present embodiment. [Figure 4] FIG. 4 is a diagram showing an example of a result of a clustering process performed by a classification unit 102 according to the present embodiment. [Diagram 5] FIG. 4 is a diagram showing an example of a result of a clustering process performed by a classification unit 102 according to the present embodiment. [Figure 6] 10 is a flowchart showing an example of the flow of optimization processing for the arrangement of water level gauges by the server 10 according to the present embodiment. [Figure 7] FIG. 11 is a block diagram showing an example of a functional configuration of a server 10 according to a second embodiment of the present invention. [Figure 8]FIG. 2 is a diagram showing an overview of a process for calculating a predicted water level value by a learning engine 112 according to this embodiment. [Figure 9] 10 is a flowchart showing an example of a flow of a prediction process for a water level in a manhole by the server 10 according to the present embodiment. [Figure 10] 1 is a graph for explaining an example of the present embodiment. [Figure 11] FIG. 2 is an explanatory diagram for explaining an example of a hardware configuration of the server 10 according to the present embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In this specification and drawings, components having substantially the same functional configuration are The same reference numerals are used to denote the same components, and duplicate explanations will be omitted.

[0012] In addition, in this specification and drawings, the representative manholes described later are referred to as representative manholes 4A and Other manholes are classified as necessary, such as manhole 4B. However, if there is no need to distinguish between manholes, we will simply refer to each manhole as a manhole. It is called Rule 4.

[0013] One embodiment of the present invention is an information processing system capable of efficiently predicting the water level in a manhole. Such a system can, for example, predict the water level in a manhole and It can be used to predict flood damage caused by overflow onto the ground. As a storm water management system, the prior art is a system that monitors the weather conditions of the area and the short-term The system combines a rainfall forecast model for the monitored area with a stormwater runoff analysis model. However, such systems are subject to changes in topography and urban development. However, due to construction work, the accuracy of the runoff analysis model cannot be guaranteed. Although it is suitable for static analysis such as hazard maps, it is not suitable for real-time analysis. It is difficult to predict water levels. To achieve real-time monitoring, There is a demand for technology that can continuously predict the water level in the hole.

[0014] However, conventional systems for measuring water levels in sewer pipes in real time are One example is a manhole called a smart manhole, which is a manhole that combines a water level gauge and a wireless transmission system. Therefore, it is necessary to develop a system for all the manholes. Therefore, the arrangement of the water level meter according to one embodiment of the present invention is In this optimization method, the water level gauge is installed in which manhole among the manholes installed in a given area. The clustering method is used to optimize the placement of the The temporal changes in water levels in the manholes included in the cluster are It refers to something that is similar. Therefore, in the same cluster, the water level of other manholes can be calculated simply. This allows the water level to be estimated with a minimum water level meter (which is expensive because it has a wireless transmission function). This arrangement makes it possible to cover the expected flood area. The system is trained in advance using time-series water level data obtained from the installed water level gauge and short-term rainfall forecast information. By inputting the data into the water level prediction engine, the predicted water level of the manhole can be output. This makes it possible to predict flooding in expected flood areas with high accuracy using a limited number of water level gauges. It becomes.

[0015] FIG. 1 is an explanatory diagram for explaining an overview of an information processing system according to this embodiment. An information processing system 1 according to the embodiment includes a server 10, as shown in FIG. The information processing system 1 according to the present embodiment, for example, via another server 20 or an information terminal 30 It is possible to use it as such.

[0016] The server 10, another server 20, and an information terminal 30 are connected to each other via a network. It has been done.

[0017] The server 10 according to this embodiment is configured to estimate a water level in a water level prediction target area 2 that includes a plurality of manholes 4. Specifically, the server 10 is a server for predicting the water level of each manhole 4 in the The water level prediction area 2 includes a plurality of manholes 4 provided in a sewer pipe 3. The past water level data of several manholes 4 (representative manhole 4A in the example shown in Figure 1) The acquired data is input to a learning engine, and the manhole output from the learning engine is The server 10 outputs the prediction information regarding the water level of the water tank 4. The information is output to the server 20 or the information terminal 30. The type of the sewer pipe 3 is not particularly limited. It may be a storm water pipe, a sewage pipe, or a combined sewer pipe.

[0018] The server 10 is a hardware device such as a computer having a hardware configuration described later. The hardware that realizes the server 10 and the manner in which the hardware is provided There are no specific limitations (e.g., whether it is a stationary server or an edge device processor) The present invention can be realized by one or more computers or servers. The same is true for the hardware that realizes the information terminal 30. It may be a computer or server, or a smartphone, mobile phone, tablet, It may be realized by any terminal such as a laptop computer or a game console.

[0019] The other server 20 and the information terminal 30 obtain, for example, prediction information regarding the water level in the manhole. This is an example of an information processing device used to acquire and utilize other servers 20 and information terminals 3. 0 may be held by, for example, a national or local authority with jurisdiction over sewerage. Such forecast information may be used by general users, for example. Take measures to prevent flooding in the predicted area2 and inform residents of the possibility of flooding. It can be used to notify citizens, etc.

[0020] As described later, the server 10 may also perform other functions, such as water level prediction. Of the manholes 4 in the area 2, which manholes 4 should have water level gauges permanently installed? If there are many manholes in the water level prediction target area 2, the water level prediction target area 2 may have a function of determining whether or not the manholes are in a large amount. If there are four water levels, installing water level gauges in all of them is not feasible due to cost and equipment restrictions. On the other hand, if a water level gauge is not installed in an appropriate manhole4, In some cases, it may be difficult to predict the water level of the manhole 4. The server 10 selects which of the limited water level gauges among the manholes 4 provided in the water level prediction target area 2. This allows the installation of the water level gauge to be optimized. Within these constraints, it becomes possible to predict the water level in manhole 4 with greater accuracy.

[0021] In the first embodiment, an example of a method for optimizing the arrangement of water level gauges by the server 10 will be described below. In the second embodiment, a method for predicting the water level in a manhole by the server 10 will be described. An example of the above will be described. These embodiments may be realized by the same server 10. However, they may be implemented by different servers.

[0022] FIG. 2 is a block diagram showing an example of the functional configuration of the server 10 according to the first embodiment of the present invention. As shown in FIG. 2, the server 10 according to the present embodiment includes a data acquisition unit 101 and a classification unit. The water level / rainfall database includes a water level prediction unit 102, a determination unit 103, and a water level / rainfall DB (database) 104.

[0023] The data acquisition unit 101 acquires data on the water level of each of a plurality of manholes provided in a predetermined area. The data acquisition unit has a function to acquire each water level data acquired by the water level gauge. 101 retrieves water level data accumulated in the past from the water level / rainfall DB 104 or an external server. The water level data referred to here is time-series water level data. Time-series water level data is water level data acquired continuously or intermittently for several months, for example.

[0024] The water level data is collected from a manhole 4 installed within the water level prediction area 2. The type of water level gauge and the manner of installation are not particularly limited. Any device capable of obtaining the water level in a manhole continuously or intermittently may be used.

[0025] The data acquisition unit 101 may also acquire data on the amount of precipitation related to the manhole. The precipitation data for the manholes is, for example, the water level prediction target where the manhole 4 is located. Past precipitation data provided by organizations such as the Japan Meteorological Agency in the region to which Region 2 belongs. Alternatively, the data may be precipitation data based on a water gauge installed in the water level prediction target area 2. stomach.

[0026] The classification unit 102 classifies multiple man- The classification unit 102 has a function of classifying holes into one or more clusters. A method such as classification is used to classify the manholes 4 into one or more clusters. Here, the classifier 102 classifies one or more substitutes based on the similarity by hierarchical clustering. The surface manhole is determined for each cluster. The similarity is, for example, the similarity of the change in water level over time. In other words, if there are some similarities in the time series changes of water levels, The threshold is set as the approximate value in the clustering process. The clustering method is not particularly limited, but may be determined as appropriate within a cluster or among clusters. A method that allows for evaluation of distances outside the raster is preferred.

[0027] FIG. 3 is an example of the clustering process of the manhole 4 by the classification unit 102 according to this embodiment. The classification unit 102 classifies the water level of the manhole 4 acquired by the data acquisition unit 101. For each data, clustering processing is performed using techniques such as structured clustering. As a result, for example, in the water level prediction target area 2, multiple clusters 5 (shown in FIG. 3) are generated. In this example, they can be classified into four clusters5).

[0028] 4 and 5 show the results of the clustering process performed by the classification unit 102 according to this embodiment. FIG. 1 is a diagram showing an example. For example, 15 manholes exist in the water level prediction target area 2. In this case, from the results of Figures 4 and 5, manholes 4, 5, and 13 are different from other manholes. As shown in Figure 5, the clusters are independent of other For example, the manholes can be classified into multiple clusters. The clusters outside of this one are the cluster of manholes number 9 and cluster A. Furthermore, in Cluster A, the cluster of manholes number 6 and C Cluster A1 and Cluster A2. The number of water level gauges that can be installed can be set appropriately. For example, as described later, The granularity of the cluster may be determined based on an upper limit on the number, etc.

[0029] The determining unit 103 determines one or more representative manholes from the plurality of manholes included in the one or more clusters. For example, the determining unit 103 has a function of determining a node for each cluster as shown in FIG. Based on the result of the clustering by the classification unit 102, one or more representative manholes 4 are classified. Determine A for each cluster 5.

[0030] For example, as shown in Figs. 4 and 5, the representative manhole 4A where the water level gauge should be installed is , manholes 4, 5, and 13 may be determined. Manholes 4, 5, and 13 Since the manholes are clustered separately from other manholes, the water levels of these manholes can be compared with those of other manholes. It is difficult to estimate the water level from the results of manholes. These manholes are then determined as representative manhole 4A.

[0031] In addition, the determination unit 103 determines whether or not the cluster list is to be updated when an upper limit is set on the number of water level gauges to be placed. Based on the results of the sampling, a representative manhole may be determined. When the upper limit of the number of positions to be arranged is 5, the determining unit 103 determines whether the position is independent of other clusters. In addition to the manholes 4, 5, and 13, there are other clusters from the top of the structure tree. The 9th manhole, which is the most isolated from the bottom of the structure tree, and the adjacent manhole from the bottom of the structure tree The manhole number 8 that most closely encompasses the hole may be determined as the representative manhole. This allows the most effective placement to be achieved, reflecting the results of clustering. .

[0032] Similarly, if the maximum number of water level gauges is 7, in addition to the 5 above, The first manhole in Cluster A1 and the sixth manhole in Cluster A2 The first manhole and the sixth manhole may be determined as the representative manhole. The manhole is chosen to improve accuracy because it is the most isolated from other manholes. It is preferable.

[0033] Next, the optimization process for the arrangement of water level gauges by the server 10 according to this embodiment will be described. FIG. 6 shows an example of a flow of optimization processing of the arrangement of water level gauges by the server 10 according to the embodiment. 1 is a flowchart showing a process of acquiring data from the server 10 in step S101. The unit 101 acquires past water level data of a plurality of manholes. Then, in step S1 In step 03, the classification unit 102 of the server 10 performs clustering on the water level data, etc. A filtering process is performed to classify multiple manholes into one or more clusters. In step S105, the determination unit 103 of the server 10 A representative manhole in the cluster is determined.

[0034] The first embodiment of the present invention has been described above. Even if water level gauges are installed in all manholes in area 2, the water level gauges can be installed in each cluster. At least one manhole among the manholes in the facility shall be designated as a representative manhole and a water level gauge shall be installed. By placing the manholes in the water, it is possible to obtain sufficient data for predicting the manhole water level, which will be described later. This allows for an effective water level control even when there are constraints on cost, installation, etc. It will be possible to monitor and predict the amount of rainfall. By using rastering processing, it was possible to correlate changes in water level with the amount of precipitation. This makes it possible to perform optimization that takes into account the effects of precipitation.

[0035] Next, a second embodiment of the present invention will be described. FIG. 7 shows a second embodiment of the present invention. 7 is a block diagram showing an example of a functional configuration of the server 10 according to the present embodiment. The server 10 includes a data acquisition unit 111, a learning engine 112, an output unit 113, and a and a rank / rainfall DB (database) 114.

[0036] The data acquisition unit 111 acquires a manhole provided in a predetermined area (a water level prediction target area 2). The system has the function of acquiring historical water level data and forecast rainfall data for a specified area. The data acquisition unit 111 acquires, for example, the following data as data required for training the learning engine 112: Obtain historical time series water level data for manholes and precipitation data related to manholes The data on the amount of precipitation related to the manhole can be obtained by the method described in the first embodiment. The data acquisition unit 111 may acquire the same data as one of the manipulators, as described later. Historical water level data of manholes adjacent to the hole may be obtained. The predicted rainfall data may be data obtained from, for example, the Japan Meteorological Agency or other sources. The data may be appropriately predicted using rain gauges installed in a certain area. The data and rainfall data may be obtained every few minutes. The data may be stored in the water level / rainfall DB 114 or may be obtained from an external source as needed. It may also be something.

[0037] The learning engine 112 uses the acquired current water level data of the one manhole, the adjacent manholes, The water level of the reservoir and future predicted rainfall data are used as input values ​​to calculate the predicted future water level data. The learning engine 112 is an engine having a learning model that calculates an output value. The explanation will be divided into training the learning model and using it for prediction.

[0038] The data acquired during training of the learning model of the learning engine 112 is a time series of past water levels. Not only the data on past precipitation, but also the data on the number of manholes adjacent to a manhole, for example, Historical water level data for the manhole adjacent to the manhole may be included. The hole may be, for example, at least one manhole located upstream or downstream of a manhole in the sewer pipe 3. These data may be, for example, synthetic explanatory variables. These can be grouped into variables such as numbers and vectors.

[0039] Then, the actual manhole's past water level data and the nearby manhole's past water level data are The data includes the past cumulative precipitation data, and the resulting future time for the manhole from that point on. Using the water level at the points as training data, nonlinear multiple regression analysis such as support vector regression is performed. Or, machine learning such as neural networks are used to generate learning models. In addition, such learning models are updated appropriately even when used for prediction. Such training may be carried out, for example, in optimizing the arrangement of the water level gauge in the above embodiment. The system utilizes water level data and precipitation data obtained from water level gauges in each manhole. may be also possible.

[0040] A learning engine based on a learning model trained on these data Water Level Prediction Engine) is This is used to calculate the predicted water level in the manhole after (r is any positive number). 1 shows an overview of the process of calculating a predicted water level value by the learning engine 112 according to this embodiment. As shown in FIG. 8, the learning engine 112 uses the past Water level data P,t and the historical water level data of nearby manholes Pref,t and r minutes Predicted precipitation data for later x t+r Using this, the predicted value of the manhole water level after r minutes, l P,t ,r The learning engine 112 may also use past precipitation data as input values ​​to calculate Such a manhole water level prediction value may be, but need not be, a numerical value. It may be calculated every minute. Also, the past water level data of the nearby manholes may be used. may be obtained directly by a water level gauge, or may be determined using the technology of the first embodiment. The water level gauge of the representative manhole determined and the water level gauge of the neighboring manholes obtained during the clustering process are The water level may be estimated based on the correlation between the water level in the manhole and the representative manhole. Even in the latter case, the accuracy of the forecast can be improved because it reflects time-series data. It can be done.

[0041] The learning engine 112 also uses past water level data for a manhole to learn A predicted value of the water level of a manhole other than the manhole may be calculated. In this case, for example, The other manhole may be a manhole adjacent to the first manhole. By using the predicted water level of the representative manhole where the water level gauge is installed, It is also possible to predict the water level in nearby manholes that are not installed.

[0042] The output unit 113 outputs the future time of the manhole calculated by the learning engine 112. The output unit 113 has a function of outputting the predicted water level information. The predicted water level of the manhole and information based on the predicted value (e.g., when the water level in the manhole reaches a dangerous level) The output of this forecast information includes the following: For example, it may be transmitted to the server 20, the information terminal 30, or the like.

[0043] Next, the process of predicting the water level in the manhole by the server 10 according to the present embodiment will be described. FIG. 9 shows an example of a flow of a prediction process of a water level in a manhole by the server 10 according to the embodiment. 1 is a flowchart showing an example of a process for The acquisition unit 111 acquires past water level data and predicted rainfall data for a plurality of manholes. Next, in step S113, the learning engine 112 of the server 10 The data is input and future manhole predicted water level data is calculated. Then, in step S115 In the above, the output unit 113 of the server 10 outputs the predicted water level data to the manhole based on the calculated predicted water level data. The system outputs predicted water level information.

[0044] The second embodiment of the present invention has been described above. When predicting the water level of the manholes in area 2, the topographical information of the water level prediction area 2 is taken into consideration. The water level data and precipitation data of manholes can be used without using information from the sewer pipe model. By simply using the learning model based on the data, the water level of the manholes included in the water level prediction target area 2 can be predicted. In addition, the water level can be predicted for all manholes included in the water level prediction area 2. Even if a water level meter is not provided, the technology described in the first embodiment can be used, for example. If the water level data of the identified representative manhole is available, it is possible to It is also possible to predict the water level in manholes.

[0045] In the above embodiment, the water level in the manhole is predicted using the predicted rainfall data. However, the present technology is not limited to such an example. For example, the learning can be performed without using the predicted rainfall data. The learning model of the engine 112 is configurable, and the learning engine 112 uses past man-hours Using the water level data from the manhole, it is possible to calculate the predicted water level in the manhole at a future point in time. This makes it possible to predict water levels when rainfall does not need to be taken into account. In the above embodiment, the water level indicators arranged optimally by the technology according to the first embodiment are used to measure the water level. However, the present technology is not limited to such an example. For example, Even if a water level gauge is provided in the manhole, the technology according to the present embodiment can be applied to the manhole. is applicable.

[0046] In addition, the forecast precipitation data used to predict water levels is provided by agencies such as the Japan Meteorological Agency. It may be data on the amount of rainfall in the past, or it may be processed data on the amount of rainfall predicted in the past. Generally, forecast precipitation data obtained from the Japan Meteorological Agency and other meteorological data providers are indexed. It is provided at intervals (for example, every 30 minutes), so it is shorter than the interval. Therefore, for example, the data of the predicted precipitation amount is The precipitation data measured after the corresponding time is used as the predicted precipitation data. For example, the reservation for 20 minutes after 15:00 (i.e. 15:20) If there is measured precipitation data, and the data was actually obtained at 15:20, In this case, the precipitation data measured at 15:20 is used as the 15:00 forecast. It is also possible to use 20-minute forecast precipitation data. This will allow for more accurate water level forecasts. In addition, the predicted precipitation data actually input to the learning engine 112 is The value may be a value, or it may be data with a margin in the positive and negative directions from the predicted precipitation amount. For example, if the forecast precipitation is 30 mm, it is possible to choose between 50% more, 45 mm, or 50% less, 45 mm. 15mm may be used as the input value. This allows the following to be used for the water level forecast information: It is possible to provide information that takes volatility into account.

[0047] Fig. 10 is a graph for explaining an example of this embodiment. The graph shows the actual water level, and the dashed line shows the data learned from the predicted rainfall data mentioned above, which is 50% larger than the actual water level. The predicted water level of the manhole is obtained by inputting the data into the engine 112. The dashed and dotted line indicates the predicted rainfall data. The predicted water level of the manhole obtained by inputting 50% less data into the learning engine 112. As shown in the graph, the predicted rainfall data, manhole water level data, and Water level data for manholes adjacent to the manhole is collected every 5 minutes, and the predicted water level and The actual measured values ​​of the water level and the precipitation are output at 5-minute intervals. As shown in the graph in Figure 10, It can be seen that there is no significant difference between the predicted water level and the actual measured water level, and they are almost the same. In this way, the past water level data and the predicted rainfall data of the manhole are used in the learning engine described above. By using the gin, it is possible to predict the water level in the manhole with high accuracy.

[0048] The information processing described above is carried out by software and the hardware of the server 10 described below. The hardware configuration described below is realized by the cooperation of the server 20 and the information The present invention is also applicable to the information terminal 30.

[0049] FIG. 11 is a diagram illustrating an example of a hardware configuration of the server 10 according to the present embodiment. As shown in FIG. 10, the server 10 includes a control unit 11, a memory 12, a storage device, and a The system includes a storage 13, a transmission / reception unit 14, an input / output unit 15, etc., which are electrically connected to each other via a bus 16. are electrically connected.

[0050] The control unit 11 controls the overall operation of the server 10 and controls data transmission and reception between each element. It is a computing device that performs information processing required for control, application execution, and authentication processing. For example, the control unit 11 includes a CPU (Central Processing Unit) and and / or GPU (Graphics Processing Unit) for Executes programs for this system stored in storage 13 and deployed in memory 12 Then, the various information processes described above are carried out.

[0051] Memory 12 is a DRAM (Dynamic Random Access Memory The main memory is made up of volatile storage devices such as flash memory and HDD (Hard Disk Drive). and auxiliary storage configured with non-volatile storage devices such as a disk drive. 12 is used as a work area for the control unit 11 and is executed when the server 10 is started. The BIOS (Basic Input / Output System) and each Stores species setting information, etc.

[0052] The storage 13 stores various programs such as application programs. A database that stores data used for processing may be constructed in storage 13. stomach.

[0053] The transceiver 14 connects the server 10 to a network and / or a blockchain network. The transmitter / receiver 14 is connected to the network. It is equipped with a short-range communication interface (Bluetooth Low Energy). This is also fine.

[0054] The input / output unit 15 includes information input devices such as a keyboard, a microphone, and a mouse, and a display. and output devices such as displays and speakers.

[0055] A bus 16 is commonly connected to each of the above elements, and transmits, for example, address signals, data signals, and The control signal is transmitted.

[0056] The information processing performed by an external device such as the information processing device 10 may be performed on-premise. The external device in the cloud format may be, for example, SaaS (Software as a Service), cloud computing The above-described functions and processes may be provided in any form.

[0057] In the above embodiment, the information processing device 10 performs various storage and control operations. Instead of 0, multiple external devices may be used. That is, various information and programs The data may be stored in a distributed manner across multiple external devices using block chain technology or the like.

[0058] The above embodiment is not limited to the information processing system 1, and may be an information processing method or an information processing The information processing method may include each step of the information processing system 1. The information processing program is installed in at least one computer, and each stage of the information processing system 1 is Execute the step.

[0059] Finally, various embodiments of the present disclosure have been described, but these are presented as examples. and are not intended to limit the scope of the invention. It is possible to embody various forms, and various omissions may be made without departing from the spirit of the invention. The embodiments and modifications thereof are within the scope and spirit of the invention. The inventions described in the claims and their equivalents are included in the scope of the invention. be.

[0060] The following embodiments are also included in the scope of the present invention. (Item 1) A method for optimizing the arrangement of water level gauges, comprising the steps of: Each water level measured by a water level gauge installed in each of a plurality of manholes in a specified area A data acquisition step of acquiring individual water level data; By performing clustering for each of the water level data, the plurality of manholes are A classification step of classifying into one or more clusters; From the manholes included in the one or more clusters, one or more representative manholes are selected. A determination step for each raster; A method for optimizing the placement of water level gauges, including: (Item 2) A method for optimizing the arrangement of the water level gauge according to item 1, In the determining step, the one or more representatives are determined based on the result of the clustering. Determine the manholes for each cluster. How to optimize water level gauge placement. (Item 3) A method for optimizing the arrangement of water level gauges according to item 2, The clustering used in the classification step is hierarchical clustering; In the determining step, the first cluster is determined based on the closeness by the hierarchical clustering. The representative manholes are determined for each cluster. How to optimize water level gauge placement. (Item 4) A method for optimizing the arrangement of water level gauges according to item 2 or 3, comprising: In the determination step, if an upper limit on the number of the water level gauges to be arranged is set, determining the representative manhole based on the result of the clustering; How to optimize water level gauge placement. (Item 5) A method for optimizing the arrangement of a water level gauge according to any one of items 1 to 4, In the data acquisition step, data on the amount of precipitation relating to the manhole is further acquired. death, In the classification step, the precipitation data and each of the water level data are used to classify the Steering, How to optimize water level gauge placement. (Item 6) A method for optimizing the arrangement of a water level gauge according to any one of items 1 to 5, The water level data is time-series water level data. How to optimize water level gauge placement. (Item 7) A method for predicting a water level in a manhole, comprising the steps of: A time series of past water level data of a manhole provided in a specified area and a water level data of a manhole provided in the specified area A past data acquisition step of acquiring the predicted rainfall data; The acquired past time series water level data of the one manhole and the predicted rainfall data By inputting the above information into a predetermined engine, the water level of the manhole at a future time point can be predicted. an output step of outputting the measurement information; A method for predicting water levels in a manhole, including: (Item 8) Item 7: A method for predicting a water level in a manhole according to the present invention, In the past data acquisition step, other manholes adjacent to the one manhole We further obtained historical time series water level data of In the output step, the past time series water level data of the other manholes is output to the Input to a specific engine, How to predict manhole water levels. (Item 9) Item 7 or 8, a method for predicting a water level in a manhole according to item 7 or 8, The past time series water level data is provided in a manhole different from the first manhole. The estimated water level data is estimated based on the water level data acquired by the water level gauge. How to predict manhole water levels. (Item 10) A method for predicting a water level in a manhole according to any one of items 7 to 9, The predetermined engine includes at least the past time series water level data of the one manhole. Based on a learning model obtained by learning using the past rainfall data related to the target area, It is a fast engine. How to predict manhole water levels. (Item 11) Item 11. A method for predicting a water level in a manhole according to item 10, The predetermined engine further records the history of other manholes adjacent to the one manhole. Obtained by learning from time-series water level data. How to predict manhole water levels. (Item 12) A method for predicting a water level in a manhole according to any one of items 7 to 11, In the output step, together with the one manhole or Instead, the prediction of the water level of the other manholes included in the target area at a future time Output the measurement information. How to predict manhole water levels. (Item 13) A manhole water level prediction system, comprising: The past time series water level data of one manhole included in the target area and the water level data of the specified area A past data acquisition unit that acquires forecast rainfall data; The acquired past time series water level data of the one manhole and the predicted rainfall data By inputting the above information into a predetermined engine, the water level of the manhole at a future time point can be predicted. an output unit for outputting the measurement information; A manhole water level prediction system, including: [Explanation of symbols]

[0061] 10 Server 20 Servers 30 Information terminal 101 Data Acquisition Section 102 Classification Department 103 Decision Section 111 Data Acquisition Section 112 Learning Engine 113 Output section

Claims

1. A method for optimizing the arrangement of water level gauges, comprising the steps of: Each water level measured by a water level gauge installed in each of a plurality of manholes in a specified area A data acquisition step of acquiring individual water level data; By performing clustering for each of the water level data, the plurality of manholes are A classification step of classifying into one or more clusters; From the manholes included in the one or more clusters, one or more representative manholes are selected. A determination step for each raster; A method for optimizing the placement of water level gauges, including:

2. 2. The method for optimizing the arrangement of water level indicators according to claim 1, comprising the steps of: In the determining step, the one or more representatives are determined based on the result of the clustering. Determine the manholes for each cluster. How to optimize water level gauge placement.

3. 3. The method for optimizing the arrangement of water level indicators according to claim 2, comprising the steps of: The clustering used in the classification step is hierarchical clustering; In the determining step, the first cluster is determined based on the closeness by the hierarchical clustering. The representative manholes are determined for each cluster. How to optimize water level gauge placement.

4. A method for optimizing the arrangement of water level gauges according to claim 2 or 3, comprising the steps of: In the determination step, if an upper limit on the number of the water level gauges to be arranged is set, determining the representative manhole based on the result of the clustering; How to optimize water level gauge placement.

5. 2. The method for optimizing the arrangement of water level indicators according to claim 1, comprising the steps of: In the data acquisition step, data on the amount of precipitation relating to the manhole is further acquired. death, In the classification step, the precipitation data and each of the water level data are used to classify the Steering, How to optimize water level gauge placement.

6. 2. The method for optimizing the arrangement of water level indicators according to claim 1, comprising the steps of: The water level data is time-series water level data. How to optimize water level gauge placement.

7. A method for predicting a water level in a manhole, comprising the steps of: A time series of past water level data of a manhole provided in a specified area and a water level data of a manhole provided in the specified area A past data acquisition step of acquiring the predicted rainfall data; The acquired past time series water level data and predicted rainfall data of the one manhole are By inputting the information into a predetermined engine, the predicted water level of the manhole at a future time point is obtained. an output step of outputting information; A method for predicting water levels in a manhole, including:

8. A method for predicting a water level in a manhole according to claim 7, In the past data acquisition step, other manholes adjacent to the one manhole We further obtained historical time series water level data of In the output step, the past time series water level data of the other manholes is output to the Input to a specific engine, How to predict manhole water levels.

9. A method for predicting a water level in a manhole according to claim 7 or 8, comprising: The past time series water level data is provided in a manhole different from the first manhole. The estimated water level data is estimated based on the water level data acquired by the water level gauge. How to predict manhole water levels.

10. A method for predicting a water level in a manhole according to claim 7 or 8, comprising: The predetermined engine includes at least the past time series water level data of the one manhole. Based on a learning model obtained by learning using the past rainfall data related to the target area, It is a fast engine. How to predict manhole water levels.

11. A method for predicting a water level in a manhole according to claim 10, The predetermined engine further records the history of other manholes adjacent to the one manhole. Obtained by learning from time-series water level data. How to predict manhole water levels.

12. A method for predicting a water level in a manhole according to claim 7 or 8, comprising: In the output step, together with the one manhole or Instead, the prediction of the water level of the other manholes included in the target area at a future time Output the measurement information. How to predict manhole water levels.

13. A manhole water level prediction system, comprising: The past time series water level data of one manhole included in the target area and the water level data of the specified area A past data acquisition unit that acquires forecast rainfall data; The acquired past time series water level data and predicted rainfall data of the one manhole are provided. By inputting the information into a specified engine, the predicted water level of the manhole at a future time point is obtained. an output unit that outputs A manhole water level prediction system, including:

Images