Water supply management device, water supply management method, program, and water supply management system

Abstract

To provide a technology for assisting in accurate water pressure forecast for a water supply pipe.SOLUTION: A water supply management device includes: a water pressure information acquiring unit for acquiring water pressure information over time at each of observation points on the water supply pipe to which water is supplied by a water supply facility; a clustering execution unit for dividing the water pressure information into a plurality of groups; a forecast model generation unit for generating a water pressure forecast model using the water pressure information; and a demand forecast unit for predicting the water pressure of the water supply at the observation point by estimating the group to which the water pressure of the forecast date at the observation point belongs using the water pressure forecast model.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a waterworks management device, a waterworks management method, a program, and a waterworks management system. [Background technology]

[0002] Patent Document 1 discloses technology related to a water supply device connected to a water supply pipe. The water supply device determines whether or not a wait-and-see operation period is necessary and the length of the period during which the pump operates in a wait-and-see operation, depending on the frequency of water use. The control device of the water supply device refers to the set data during time periods when water use is low and determines a short wait-and-see operation period. Furthermore, during time periods when water use is high, the control device refers to the set data and determines a long wait-and-see operation period. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2021-085369 Summary of the Invention [Problem to be solved by the invention]

[0004] In waterworks businesses, equipment is operated according to the demand of each customer to supply water. It is cumbersome for workers to monitor demand and adjust operation settings as needed based on the monitoring results when supplying water. Meanwhile, when equipment is operated based on a preset operation schedule, if the operation schedule does not match the actual situation, it can lead to supply shortages due to insufficient operation, or to poor energy efficiency and equipment deterioration due to over-operation. The technology disclosed in Patent Document 1 determines the downtime period after pump use, but does not predict demand before operation.

[0005] The present invention has been made in consideration of the above points, and aims to provide a technology that supports highly accurate prediction of water pressure in water supply piping. [Means for solving the problem]

[0006] The present application includes a number of means for solving the above problems, examples of which are as follows.

[0007] In order to solve the above problem, one embodiment of the water supply management device of the present invention is characterized by comprising a water pressure information acquisition unit that acquires water pressure information over time for each observation point of a water supply pipe supplied by a water supply facility; a clustering execution unit that divides the water pressure information into multiple groups; a prediction model generation unit that generates a water pressure prediction model using the water pressure information; and a demand prediction unit that predicts the water pressure of the water supply at an observation point by using the water pressure prediction model to estimate the group to which the water pressure at the observation point on a predicted day belongs.

[0008] The water pressure information may be information regarding water pressure over time on a daily basis, and is associated with attribute information including the day of the week and weather on the observation date, and the water pressure prediction model may be characterized in that it uses at least one of the attribute information as an explanatory variable.

[0009] The water pressure prediction model may be characterized in that past water pressure information at the observation point to be predicted is used as an explanatory variable.

[0010] The water supply management device may include an equipment schedule generation unit that generates an operation schedule for the water supply equipment that supplies water to the water supply piping at multiple observation points, and the equipment schedule generation unit may be characterized in that it generates the operation schedule for the water supply equipment that supplies water to the multiple observation points so that the water pressure information at the multiple observation points on the predicted date satisfies specified conditions.

[0011] The water supply equipment may be a pump that supplies water to the water supply pipes, and the equipment schedule generation unit may generate the operation schedule for the pump so that the water pressure of the multiple water supply pipes is always above a predetermined value.

[0012] The facility schedule generating unit may select one operation schedule from a plurality of patterns of operation schedule candidates based on the prediction by the demand forecasting unit.

[0013] The demand forecasting unit may be characterized by having a prediction accuracy determination unit that determines predicted day water pressure information, which is the water pressure over time on the predicted day, and after the predicted day on which the demand forecasting unit predicted the water pressure has passed, compares the water pressure information for the predicted day acquired by the water pressure information acquisition unit with the predicted day water pressure information predicted by the demand forecasting unit, and determines whether the comparison result satisfies specified conditions.

[0014] The clustering execution unit may cluster the water pressure information at each observation point, the demand forecasting unit may determine predicted day water pressure information for each observation point, which is the water pressure over time on the predicted day, and the water pressure forecasting model may use the predicted day water pressure information of other observation points that are pre-associated as explanatory variables for the prediction of a certain observation point.

[0015] In addition, in order to solve the above-mentioned problems, another aspect of the present invention provides a water supply management method, which is characterized by comprising a water pressure information acquisition step for acquiring water pressure information over time for each observation point of a water supply pipe supplied by a water supply facility; a clustering execution step for dividing the water pressure information into multiple groups; a prediction model generation step for generating a water pressure prediction model using the water pressure information; and a demand prediction step for predicting the water pressure of the water supply at an observation point by using the water pressure prediction model to estimate the group to which the water pressure at the observation point on a predicted day belongs.

[0016] In addition, in order to solve the above-mentioned problems, a program according to another aspect of the present invention is a program that causes a computer processing unit to execute a water supply management method, and is characterized by executing a water pressure information acquisition step that acquires water pressure information over time for each observation point of a water supply pipe supplied by a water supply facility, a clustering execution step that divides the water pressure information into multiple groups, a prediction model generation step that generates a water pressure prediction model using the water pressure information, and a demand prediction step that predicts the water pressure of the water supply at a certain observation point by using the water pressure prediction model to estimate the group to which the water pressure at the observation point on a predicted day belongs.

[0017] In addition, in order to solve the above-mentioned problems, another aspect of the present invention provides a water supply management system comprising a water supply management device, water pressure sensors installed at each observation point of the water supply pipes to which water is supplied by the water supply equipment, and a water supply equipment control device that controls the water supply equipment, wherein the water pressure sensors measure the water pressure of water passing through the water supply pipes, and the water supply management device comprises: a water pressure information acquisition unit that acquires time-dependent water pressure information at each observation point generated using the water pressure measured by the water pressure sensors; a clustering execution unit that divides the water pressure information into multiple groups; a prediction model generation unit that generates a water pressure prediction model using the water pressure information; and a demand prediction unit that predicts the water pressure of the water supply at a certain observation point by using the water pressure prediction model to estimate the group to which the water pressure at the observation point on a predicted day belongs, and the water supply equipment control device comprises a processing unit that controls the water supply equipment to supply water using the water pressure predicted by the demand prediction unit. [Effects of the Invention]

[0018] According to the present invention, it is possible to support highly accurate prediction of water pressure in water supply piping.

[0019] Problems, configurations, and effects other than those described above will become apparent from the following description of the embodiments. [Brief explanation of the drawings]

[0020] [Figure 1] FIG. 2 is a diagram illustrating an example of functional blocks of a water supply management system. [Figure 2] FIG. 2 is a diagram illustrating an example of a data structure of a water pressure database. [Figure 3] FIG. 10 is a diagram illustrating an example of a data structure of an explanatory variable table. [Figure 4] FIG. 2 is a diagram illustrating an example of a hardware configuration of a waterworks management device. [Figure 5] 10 is a flowchart illustrating an example of a water pressure prediction model generation process. [Figure 6] Fig. 6(A) is a diagram showing an example of an overview of clustering of water pressure information. Fig. 6(A) is an example of water pressure information for multiple days at a certain observation point, Fig. 6(B) is an example of water pressure information in a certain cluster generated from the water pressure information shown in Fig. 6(A), and Fig. 6(C) is an example of water pressure information in another cluster generated from the water pressure information shown in Fig. 6(A). [Figure 7] 10 is a flowchart illustrating an example of an operation schedule determination process. [Figure 8] 10 is a flowchart illustrating an example of a prediction accuracy verification process. DETAILED DESCRIPTION OF THE INVENTION

[0021] Hereinafter, an example of an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a diagram showing an example of functional blocks of a water supply management system 1. The water supply management system 1 has a water supply management device 10, a water pressure sensor 20, and a pump control device 30, which are communicably connected to each other via a network N. The water supply management device 10 is a device managed by, for example, a local government that operates a water supply business. The water supply management device 10 is, for example, a server computer or a PC (Personal Computer).

[0022] The water supply management device 10 comprises a processing unit 110, a memory unit 120, an input unit 130, an output unit 140, and a communication unit 150. The processing unit 110 comprehensively controls the entire water supply management device 10. The memory unit 120 stores information necessary for processing by the processing unit 110. The input unit 130 accepts information input to the water supply management device 10 from an input device connected via an input IF 14, which will be described later. The output unit 140 outputs information stored in the water supply management device 10 from an output device connected via an output IF 15, which will be described later. The communication unit 150 controls the transmission and reception of information to and from other information processing devices connected so as to be able to communicate via a network N.

[0023] The processing unit 110 includes a water pressure information acquisition unit 111, a clustering execution unit 112, a prediction model generation unit 113, a demand forecasting unit 114, an equipment schedule generation unit 115, and a prediction accuracy determination unit 116. The water pressure information acquisition unit 111 acquires water pressure information for water supply pipes supplied by water supply equipment managed by a waterworks operator. The water pressure information is information detected by the water pressure sensor 20 and is information regarding the water pressure over time at each observation point. For example, the water pressure information acquisition unit 111 acquires water pressure information for one day as one data item. The water pressure information is associated with attribute information including the season, day of the week, and weather of the observation date.

[0024] The clustering execution unit 112 divides the water pressure information into multiple groups. The clustering execution unit 112 groups the information for each observation point so that similar pieces of water pressure information belong to the same group. The clustering method is not limited, and any of non-hierarchical clustering such as the k-means method, and hierarchical clustering such as the group average method, Ward's method, shortest distance method, or longest distance method may be used.

[0025] The prediction model generation unit 113 generates a water pressure prediction model using the water pressure information. The water pressure prediction model is generated by a water pressure prediction model generation process described below. As will be described in detail later, the water pressure prediction model is used to predict the water pressure over time on the prediction day using at least one of the attribute information associated with the water pressure information, such as the season, day of the week, and weather of the observation day, as an explanatory variable.

[0026] The demand forecasting unit 114 uses the water pressure prediction model to estimate the group to which the water pressure at a certain observation point on a prediction day belongs, thereby predicting the water pressure of the water supply at the observation point. The demand forecasting unit 114 generates predicted day water pressure information, which is the water pressure over time on the prediction day, for each observation point.

[0027] The equipment schedule generating unit 115 generates an operation schedule for waterworks equipment that supplies water to waterworks pipes at multiple observation points. Below, an example of generating an operation schedule for a pump that supplies water to waterworks pipes will be explained as waterworks equipment. The equipment schedule generating unit 115 generates a pump operation schedule so that water pressure information at multiple observation points supplied by one pump satisfies a predetermined condition.

[0028] The prediction accuracy determination unit 116 determines the accuracy of the prediction after the predicted day has passed on which the water pressure was predicted by the demand prediction unit 114. The prediction accuracy determination unit 116 compares the predicted day water pressure information generated by the demand prediction unit 114, with the target day as the predicted day, with the actual water pressure information for the target day acquired by the water pressure information acquisition unit 111, and determines whether the comparison result satisfies a predetermined condition.

[0029] The storage unit 120 stores a water pressure database 121 and an explanatory variable table 122. The water pressure database 121 includes water pressure information in which the actual water pressure values measured at each observation point are recorded in chronological order, and attribute information for the observation date. The explanatory variable table 122 is a table in which explanatory variables used for water pressure prediction are associated with each observation point.

[0030] The water pressure sensor 20 is a device that measures the water pressure in the water supply pipes using a predetermined method. A signal indicating the water pressure measured by the water pressure sensor 20 is transmitted to the water supply management device 10. In this embodiment, the observation point indicates the position on the water supply pipe where the water pressure sensor 20 is installed.

[0031] The pump control device 30 is an information processing device that controls pumps managed by a waterworks operator. The pump control device 30 includes a processing unit 310, a memory unit 320, an input unit 330, an output unit 340, and a communication unit 350. The processing unit 310 comprehensively controls the entire pump control device 30. The processing unit 310 controls the operation of the pump based on the operation schedule received from the waterworks management device 10. In other words, the processing unit 310 controls the pump so that water is supplied based on the operation schedule predicted by the demand prediction unit 114.

[0032] The storage unit 320 stores information necessary for processing by the processing unit 310. The input unit 330 accepts information input to the pump control device 30 from an input device connected via an input IF (not shown). The output unit 340 outputs information stored in the pump control device 30 from an output device connected via an output IF (not shown). The communication unit 350 controls the transmission and reception of information to and from other information processing devices connected so as to be able to communicate via the network N.

[0033] 2 is a diagram showing an example of the data structure of the water pressure database 121. The water pressure database 121 includes, for example, an observation point ID, a pump ID, water pressure information, observation date, season, weather, day of the week, and a holiday flag. The observation point ID is identification information that identifies the water pressure observation point. The pump ID is identification information that identifies the pump that supplies water to the observation point.

[0034] Water pressure information is information that indicates the water pressure measured at an observation point. More specifically, water pressure information is information that includes multiple observation times within a certain period (e.g., one day) and the water pressure values at the observation times. For example, water pressure information includes observation times at equal intervals (e.g., one-minute intervals) and the water pressure values at those observation times. As mentioned above, in this example, the water pressure information indicates the water pressure over time for one day.

[0035] The observation date is information that identifies the date on which the water pressure information was obtained. The season is information that identifies the season of the observation date. The weather is information that identifies the weather of the observation date. The day of the week is information that identifies the day of the week of the observation date. The holiday flag is information that indicates whether the observation date is a holiday. The season, weather, day of the week, and holiday flag can be said to be attribute information of the observation date.

[0036] Additionally, the attribute information included in the water pressure database 121 is not limited to the example shown in this figure. For example, it may include a Goto day flag indicating whether or not a day ending in "5" or "10" is a "Goto day."

[0037] 3 is a diagram showing an example of the data structure of the explanatory variable table 122. The explanatory variable table 122 is information in which items of explanatory variables used for water pressure prediction are associated with each observation point that is the management target of the waterworks management device 10.

[0038] Attribute information of the observation date, such as season, weather, day of the week, and holiday flag, is used as explanatory variables for water pressure prediction. Additionally, explanatory variables specific to the observation point, such as whether or not a facility near the observation point is in operation, can be used for water pressure prediction. Information indicating which explanatory variable items are used when predicting water pressure at an observation point is associated with the explanatory variable table 122.

[0039] 4 is a diagram showing an example of the hardware configuration of the water supply management device 10. The water supply management device 10 includes a calculation device 11, a memory 12, an external storage device 13, an input IF (Interface) 14, an output IF 15, and a communication IF 16, and each component is connected by a bus.

[0040] The arithmetic device 11 is a arithmetic device such as a CPU (Central Processing Unit), and in the waterworks management device 10, which executes processing according to a program recorded in the memory 12 or the external storage device 13, processing is performed by the arithmetic device 11 which operates according to a program read onto the memory 12 or the external storage device 13. The processing unit 110 realizes each function by the arithmetic device 11 executing the program.

[0041] The memory 12 is a storage device such as RAM (Random Access Memory) or flash memory, and functions as a storage area from which programs and data are temporarily read. The external storage device 13 is a writable and readable storage medium and storage media drive, such as an HDD (Hard Disk Drive), CD-R (Compact Disc-Recordable), or DVD-RAM (Digital Versatile Disk-Random Access Memory). The functions of the storage unit 120 are realized by the memory 12 or the external storage device 13. Note that the functions of the storage unit 120 may also be realized by a storage device connected via the communication IF 16.

[0042] The input IF 14 is an interface for receiving input operations from an operator, and is connected to input devices such as a touch panel, keyboard, mouse, microphone, etc. The output IF 15 is an interface for outputting information to an output device such as an OLED (Organic Light Emitting Diode) display built into the waterworks management device 10.

[0043] The communication IF 16 is an interface for connecting the waterworks management device 10 to the network N, and is connected to a communication device such as a LAN (Local Area Network) card. The waterworks management device 10 may also have a storage medium drive device (not shown) for inputting and outputting information from portable media such as a CD (Compact Disk) or a DVD (Digital Versatile Disk).

[0044] The processing of each component of the water supply management device 10 may be executed by one piece of hardware or by multiple pieces of hardware. Also, the processing of each component of the water supply management device 10 may be realized by one program or by multiple programs.

[0045] The hardware configuration of the pump control device 30 is the same as that of the waterworks management device 10, and therefore a description thereof will be omitted.

[0046] 5 is a flowchart showing an example of the water pressure prediction model generation process. The process of this flowchart starts when the waterworks management device 10 receives, for example, an input operation indicating a generation request for the water pressure prediction model generation process. As an example, the water pressure prediction model generation process is executed every predetermined period (for example, every six months).

[0047] First, the clustering execution unit 112 determines an observation point to be classified (step S11). For example, the clustering execution unit 112 determines one observation point to be managed by the waterworks management service.

[0048] Next, the clustering execution unit 112 extracts water pressure information for the determined observation points (step S12). Specifically, the clustering execution unit 112 references the water pressure database 121 and extracts water pressure information associated with the observation point ID of the observation point determined in step S11. In other words, the water pressure information extracted by the clustering execution unit 112 is information about water pressure measured on multiple observation dates at the target observation point.

[0049] Next, the clustering execution unit 112 determines the definition of distance to be used for clustering and clusters the water pressure information (step S13). Specifically, the clustering execution unit 112 receives a specification, via an input operation by the user, of the definition of distance to be used for determining whether or not each piece of water pressure information is similar. Note that the definition of distance to be used for clustering may be a definition that has been specified in advance.

[0050] The clustering execution unit 112 also converts the water pressure information extracted in step S12 into a high-dimensional vector and performs clustering using distances based on the determined definition. For example, the time-series water pressure information for each observation date can be converted into a 24-dimensional vector with the hourly water pressure average value as the first dimension. Note that the preprocessing performed when clustering the water pressure information is not limited to conversion into a high-dimensional vector. As mentioned above, existing clustering methods can also be used.

[0051] Figure 6 is a diagram showing an example of an overview of clustering of water pressure information. Figure 6(A) is an example of water pressure information for multiple days at a certain observation point, Figure 6(B) is an example of water pressure information in a certain cluster generated from the water pressure information shown in Figure 6(A), and Figure 6(C) is an example of water pressure information in another cluster generated from the water pressure information shown in Figure 6(A).

[0052] Figure 6(A) shows water pressure information measured over time on "1 / 11," "1 / 12," "1 / 13," and "1 / 14." Figure 6(B) shows the water pressure information for "1 / 12" and "1 / 13" clustered as water pressure information belonging to the same cluster. Figure 6(C) shows the water pressure information for "1 / 11" and "1 / 14" clustered as water pressure information belonging to the same cluster.

[0053] Returning to FIG. 5 for explanation, the prediction model generation unit 113 then identifies explanatory variables and generates a water pressure prediction model (step S14). Specifically, the prediction model generation unit 113 refers to the explanatory variable table 122 and extracts explanatory variable items associated with the observation point determined in step S11. The prediction model generation unit 113 identifies explanatory variables corresponding to the extracted items. For example, the prediction model generation unit 113 can acquire explanatory variables included in the water pressure database 121 extracted in step S12.

[0054] The prediction model generation unit 113 generates a water pressure prediction model using the identified explanatory variables, with the water pressure information contained in each cluster as the objective variable. For example, the prediction model generation unit 113 can generate a water pressure prediction model using known techniques such as multiple regression analysis, decision trees, logistic regression, and random forests.

[0055] Note that water pressure information may be related to past water pressure information at the same observation point. Therefore, the prediction model generation unit 113 may use past water pressure information at the observation point to be predicted as an explanatory variable. In this case, for example, the explanatory variable table 122 contains information specifying how far back in time, based on the observation date, past water pressure information should be used as an explanatory variable for a given observation point. The prediction model generation unit 113 uses the values contained in the explanatory variable table 122 to specify the past water pressure information to use as an explanatory variable.

[0056] Furthermore, water pressure information may be related to water pressure information predicted at other observation points. Therefore, when generating a prediction model for a certain observation point (e.g., point A), the prediction model generation unit 113 can use predicted day water pressure information, which is water pressure information predicted at another previously associated observation point (e.g., point B), as an explanatory variable. In this case, the demand prediction unit 114 performs prediction for the observation point (point B) used as an explanatory variable before predicting the observation point (point A) that will use it.

[0057] Next, the clustering execution unit 112 determines whether all observation points have been processed (step S15). The clustering execution unit 112 determines whether there are any observation points that have not yet been determined as processing points in step S11 among the observation points that are managed by the water supply management service.

[0058] If the clustering execution unit 112 determines that not all observation points are to be processed ("NO" in step S15), the processing unit 110 moves the process to step S11.

[0059] If the clustering execution unit 112 determines that all observation points have been processed (YES in step S15), the processing unit 110 ends the processing of this flowchart.

[0060] 7 is a flowchart showing an example of an operation schedule determination process. The process of this flowchart is executed, for example, daily. Below, an example will be described in which the waterworks management device 10 determines the operation schedule for the next day.

[0061] First, the demand forecasting unit 114 determines a pump to be processed (step S21). Specifically, the demand forecasting unit 114 determines one of the pumps managed by the waterworks management device 10 as the pump to be processed.

[0062] Next, the demand forecasting unit 114 extracts the observation points to which the pumps will supply water (step S22). Specifically, the demand forecasting unit 114 refers to the water pressure database 121 and extracts the observation point IDs associated with the pumps determined in step S21.

[0063] Next, the demand forecasting unit 114 acquires attribute information for the prediction date (step S23). Specifically, the demand forecasting unit 114 refers to the explanatory variable table 122 to identify items to be used in predicting the water pressure at the observation point extracted in step S22, and acquires attribute information that serves as the values of the items. The demand forecasting unit 114 acquires attribute information for the prediction date, including, for example, information indicating the next day's season, weather, day of the week, and whether it is a public holiday. Furthermore, for example, the demand forecasting unit 114 acquires, as the attribute information for the observation point, water pressure information from the water pressure information at the observation point for a period going back the period identified in the explanatory variable table 122. The demand forecasting unit 114 may acquire attribute information using a web page or the like displayed by another information processing device connected via the Internet, or may acquire attribute information based on input operations from the user.

[0064] Next, the demand forecasting unit 114 predicts demand at each observation point (step S24). That is, the demand forecasting unit 114 predicts the water pressure of the water supply at each observation point by estimating the cluster to which the water pressure at that observation point on the prediction day belongs, using the water pressure prediction model for each observation point generated by the water pressure prediction model generation process. For example, the demand forecasting unit 114 estimates the cluster to which the water pressure on the prediction day belongs for each observation point, and sets a representative value of that cluster as the water pressure information for the prediction day. Note that the method for determining the representative value is not limited, and for example, the average value of the water pressure information for each of the water pressures that make up the cluster can be used.

[0065] Next, the facility schedule generation unit 115 determines the operation schedule candidate that will result in the water pressure at each observation point being equal to or greater than a predetermined value as the operation schedule to be executed (step S25). Note that, before this process begins, the pump operation schedule candidates are stored in advance in an area (not shown) of the storage unit 120. From the operation schedule candidates, the facility schedule generation unit 115 selects one operation schedule for which the predicted daily water pressure information at each observation point predicted in step S24 satisfies predetermined conditions.

[0066] For example, the facility schedule generating unit 115 may determine, as the operation schedule to be executed, a candidate operation schedule for operating the pump so that the water pressure at each observation point is always equal to or greater than a predetermined value. Setting a threshold value at each observation point so as not to impede stable use can prevent supply shortages due to insufficient operation. Furthermore, for example, the facility schedule generating unit 115 may determine, as the operation schedule to be executed, a candidate operation schedule for operating the pump so that the value indicating the gradient of the predicted daily water pressure information at each observation point does not exceed a predetermined value.

[0067] In other words, the facility schedule generating unit 115 may determine the operation schedule of the pump so that the predicted daily water pressure information at each observation point satisfies a predetermined condition. For example, the facility schedule generating unit 115 may determine the operation schedule by using a schedule learning model that has learned an operation schedule in which the predicted daily water pressure information satisfies a predetermined condition.

[0068] Next, the equipment schedule generating unit 115 determines whether all pumps have been designated as processing targets (step S26). The equipment schedule generating unit 115 determines whether all pumps to be managed by the waterworks management device 10 have been designated as processing targets in step S21.

[0069] If the equipment schedule generating unit 115 determines that not all pumps are to be processed ("NO" in step S26), the processing unit 110 shifts the process to step S21.

[0070] If the equipment schedule generating unit 115 determines that all pumps have been processed (YES in step S26), the processing unit 110 ends the processing of this flowchart.

[0071] As described above, in this embodiment, predicted day water pressure information is generated for each observation point. However, the demand forecasting unit 114 may generate one predicted day water pressure information using water pressure information from multiple observation points. In this case, the prediction model generation unit 113 can generate one integrated water pressure information from the water pressure information from the multiple observation points, cluster the integrated water pressure information, and generate a water pressure prediction model.

[0072] This embodiment makes it possible to obtain highly accurate prediction results for water pressure information. Furthermore, this embodiment makes it possible to more efficiently create operation schedules for pumps, which are waterworks facilities that supply water to multiple observation points. Furthermore, by using the operation schedules created by this embodiment, it is possible to provide water supply that is more in line with actual conditions, thereby preventing a deterioration in energy efficiency due to oversupply and a supply shortage.

[0073] Figure 8 is a flowchart showing an example of a prediction accuracy verification process. The process of this flowchart is executed at any time after the predicted date on which the water pressure was predicted in the operation schedule determination process has passed. In this process, after the predicted date on which the water pressure was predicted by the demand forecasting unit 114 has passed, the actual water pressure information for the predicted date acquired by the water pressure information acquisition unit 111 is compared with the predicted day water pressure information, and it is determined whether the comparison result satisfies a predetermined condition.

[0074] First, the prediction accuracy determination unit 116 identifies a target date and an observation point (step S31). Specifically, the prediction accuracy determination unit 116 identifies a target date and an observation point for accuracy determination. The method for identifying the target date and observation point is not limited, and for example, the prediction accuracy determination unit 116 can identify a target date and observation point for which an input operation is received from a user.

[0075] Next, the prediction accuracy determination unit 116 identifies the predicted day water pressure information, which uses the target date as the prediction date, and the actual water pressure information for the target date (step S32). Specifically, the prediction accuracy determination unit 116 identifies the predicted day water pressure information generated for the observation point identified in step S31, which uses the target date as the prediction date. The prediction accuracy determination unit 116 also identifies the water pressure information associated with the observation point and the target date in the water pressure database 121 as the actual water pressure information.

[0076] Next, the prediction accuracy determination unit 116 determines whether the comparison result satisfies a predetermined condition (step S33). For example, the prediction accuracy determination unit 116 determines that the predetermined condition is satisfied if the difference between the predicted day water pressure information and the water pressure information identified in step S32 is less than a predetermined value. Then, the prediction accuracy determination unit 116 ends the processing of this flowchart.

[0077] If the prediction accuracy determination unit 116 determines in step S33 that the conditions are not satisfied, it may output alert information specifying the target date and observation point to the user. Additionally, the prediction accuracy determination unit 116's prediction accuracy determination is not limited to a method using the difference between the predicted day water pressure information and the actual water pressure information. For example, the prediction accuracy determination unit 116 may determine whether the water pressure included in the predicted day water pressure information is always equal to or greater than a predetermined value.

[0078] As described above, by evaluating the prediction accuracy of the predicted day water pressure information, it can be used to tune the water pressure prediction model.

[0079] Although the above describes each embodiment of the present invention, the present invention is not limited to the above-described exemplary embodiment and includes various modifications. For example, the above-described exemplary embodiment has been described in detail to facilitate understanding of the present invention, and the present invention is not limited to an embodiment including all of the components described herein. Furthermore, part of the components of one exemplary embodiment can be replaced with the components of another exemplary embodiment. Furthermore, the components of another exemplary embodiment can be added to the components of one exemplary embodiment. Furthermore, part of the components of each exemplary embodiment can be added, deleted, or replaced with other components. Furthermore, some or all of the above-described components, functions, processing units, processing means, etc. may be implemented in hardware, for example, by designing them as integrated circuits. Furthermore, the control lines and information lines in the figures are only those considered necessary for explanation, and not necessarily all of them are shown. It can be assumed that almost all components are interconnected.

[0080] Furthermore, the functional configuration of the waterworks management device 10 and the pump control device 30 described above has been classified according to the main processing content for ease of understanding. The classification method and names of the components do not limit the present invention. As described above, the configuration of the waterworks management device 10 and the pump control device 30 can be classified into more components according to the processing content. Furthermore, a single component can be classified to perform even more processes. [Explanation of symbols]

[0081] 1: Water supply management system, 10: Water supply management device, 11: Computing device, 12: Memory, 13: External storage device, 14: Input IF, 15: Output IF, 16: Communication IF, 20: Water pressure sensor, 30: Pump control device, 110·310: Processing unit, 120·320: Memory unit, 130·330: Input unit, 140·340: Output unit, 150·350: Communication unit, 111: Water pressure information acquisition unit, 112: Clustering execution unit, 113: Prediction model generation unit, 114: Demand forecasting unit, 115: Equipment schedule generation unit, 116: Prediction accuracy determination unit, 121: Water pressure database, 122: Explanatory variable table,

Claims

1. a water pressure information acquisition unit that acquires time-varying water pressure information for each observation point in the water supply pipes supplied with water by the water supply facility; a clustering execution unit that divides the water pressure information into a plurality of groups; a prediction model generation unit that generates a water pressure prediction model using the water pressure information; A water supply management device characterized by comprising a demand prediction unit that predicts the water pressure of the water supply at a certain observation point by estimating the group to which the water pressure at the observation point on a predicted day belongs using the water pressure prediction model.

2. The waterworks management device according to claim 1, The water pressure information is information about water pressure over time for each day, and is associated with attribute information including the day of the week on which the observation was made, the weather on which the observation was made, and the season; A water supply management device characterized in that the water pressure prediction model uses at least one of the attribute information of the prediction date as an explanatory variable.

3. The waterworks management device according to claim 2, A water supply management device characterized in that the water pressure prediction model uses past water pressure information at the observation point to be predicted as an explanatory variable.

4. The waterworks management device according to claim 1 or 2, an equipment schedule generating unit that generates an operation schedule of the water supply equipment that supplies water to the water supply pipes at a plurality of observation points; A water supply management device characterized in that the equipment schedule generation unit generates the operation schedule of the water supply equipment that supplies water to the multiple observation points so that the water pressure information at the multiple observation points on the predicted date satisfies specified conditions.

5. The waterworks management device according to claim 4, The water supply facility is a pump that supplies water to the water supply piping, The water supply management device is characterized in that the equipment schedule generation unit generates the operation schedule of the pumps so that the water pressure of the multiple water supply pipes is always equal to or higher than a predetermined value.

6. The waterworks management device according to claim 4, The waterworks management device is characterized in that the facility schedule generation unit selects one operation schedule from a plurality of operation schedule candidate patterns based on the prediction by the demand forecasting unit.

7. The waterworks management device according to claim 1 or 2, The demand forecasting unit determines forecast day water pressure information, which is the water pressure over time on the forecast day; A water supply management device characterized by having a prediction accuracy determination unit that, after the predicted date on which the demand prediction unit predicted the water pressure has passed, compares the water pressure information for the predicted date acquired by the water pressure information acquisition unit with the water pressure information for the predicted date predicted by the demand prediction unit, and determines whether the comparison result satisfies specified conditions.

8. The waterworks management device according to claim 3, the clustering execution unit clusters the water pressure information at the observation points for each of the observation points; The demand forecasting unit determines, for each of the observation points, forecast day water pressure information, which is the water pressure over time on the forecast day; A water supply management device characterized in that the water pressure prediction model uses the predicted day water pressure information of other observation points that are previously associated as an explanatory variable for predictions at a certain observation point.

9. a water pressure information acquisition procedure for acquiring water pressure information over time at each observation point of a water supply pipe supplied with water by a water supply facility; a clustering step for dividing the water pressure information into a plurality of groups; a prediction model generation step of generating a water pressure prediction model using the water pressure information; A water supply management method characterized by comprising a demand prediction procedure for predicting the water pressure of the water supply at a certain observation point by estimating the group to which the water pressure at the observation point on a predicted day belongs using the water pressure prediction model.

10. A program that causes a processing unit of a computer to execute a water supply management method, a water pressure information acquisition procedure for acquiring water pressure information over time at each observation point of a water supply pipe supplied with water by a water supply facility; a clustering step for dividing the water pressure information into a plurality of groups; a prediction model generation step of generating a water pressure prediction model using the water pressure information; A program characterized by executing a demand prediction procedure that predicts the water pressure of the water supply system at an observation point by estimating the group to which the water pressure at the observation point on a predicted day belongs using the water pressure prediction model.

11. A water supply management system having a water supply management device, water pressure sensors installed at each observation point of a water supply pipe to which water is supplied by a water supply facility, and a water supply facility control device that controls the water supply facility, the water pressure sensor measures the water pressure in the water supply pipe; The waterworks management device includes: a water pressure information acquisition unit that acquires time-varying water pressure information at each observation point generated using the water pressure measured by the water pressure sensor; a clustering execution unit that divides the water pressure information into a plurality of groups; a prediction model generation unit that generates a water pressure prediction model using the water pressure information; a demand prediction unit that predicts the water pressure of the waterworks at an observation point by estimating the group to which the water pressure at the observation point on a prediction day belongs using the water pressure prediction model, A water supply management system, characterized in that the water supply equipment control device includes a processing unit that controls the water supply equipment so that water is supplied at the water pressure predicted by the demand prediction unit.

Images