JP2015094665A - Water leakage investigation planning device, water leakage investigation planning system, and water leakage planning method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To draw a water leakage investigation plan with a high cost-benefit performance under limited resources in a situation where there is uncertainty about water leakage.SOLUTION: A water leakage investigation planning device generates prediction model information for predicting transition in an amount of water leakage in an area on the basis of at least one information of water leakage amount information, pipeline information, and investigation/repair information; generates predicted water leakage amount information of the area on the basis of the prediction model information; and draws a water leakage investigation plan defining execution order of water leakage investigations in a plurality of areas on the basis of the predicted water leakage amount information. The predicted water leakage amount information is generated based on both prediction values of an expectation value of the amount of water leakage and the uncertainty, and draws the water leakage investigation plan using a calculated water leakage cost.

Description

  The present invention relates to a water leakage investigation planning device, and more particularly to a water leakage investigation planning device that makes a cost-effective plan.

  Claim 1 of Patent Document 1 is a leakage extraction target route extraction system for analyzing past leakage repair data collected and accumulated in a distribution pipeline network process for supplying purified water to a terminal customer, Master DB in which past leakage repair data and buried water pipe number data are accumulated, and data such as the number of repairs for each drawing number and the number of buried water pipes for each construction number, repair year, and laying year are extracted from the master DB. Based on the data extraction means and the data obtained from the data extraction means, the number of repairs and the repair ratio for each drawing number with respect to the entire distribution pipeline network are calculated, and the calculated values are digitized or highlighted for each drawing. Extract the construction number belonging to the mesh evaluation means and the drawing number with high leakage risk obtained by the mesh evaluation means, the construction number, the number of repairs in the drawing, all repairs Construction number list extraction means that lists the number of buried water supply pipes and repair ratios, and based on the extracted construction number list, leakage investigation priority evaluation that calculates leakage investigation priority based on parameters set by the user It is described as “a water leak investigation target route extraction system characterized by comprising means”.

JP 2011-59799 A

  Although the leak extraction target route extraction system of Patent Literature 1 can extract areas with a high risk of water leakage from repair information, it allocates limited resources (such as personnel) and determines which area to conduct the water leakage survey next. I can't. Moreover, the water leak investigation target route extraction system of Patent Document 1 cannot cope with the uncertainty of the water leak risk.

  SUMMARY OF THE INVENTION Accordingly, an object of the present invention is to provide a water leakage investigation plan drafting device for drafting a cost-effective water leakage investigation plan under the limited resources, even when there is uncertainty regarding water leakage.

In order to solve the above problems, for example, the configuration described in the claims is adopted. The present application includes a plurality of means for solving the above-described problems. To give an example, the present invention is a water leakage investigation plan planning device for making a water leakage investigation plan for a plurality of areas into which the water pipe network is divided.
A measurement information collection unit that collects measurement information relating to the flow rate of water from a measurement device such as a flow meter installed in the water pipe network, and a water usage storage unit that stores water usage information in the area A leakage amount estimation unit for estimating the amount of water leakage in the area based on the measurement information and the water usage information, and a pipeline for storing pipeline information including extension information of the water pipe network in the area An information storage unit, a survey / repair information storage unit for accumulating survey / repair information including the timing of water leakage investigation and pipeline repair in the area;
A prediction model learning unit that generates prediction model information for predicting a transition of the leakage amount of the area based on at least one of the leakage amount information, the pipeline information, and the investigation / repair information; and the prediction model information A leakage amount prediction unit that generates predicted leakage amount information of the area based on the information, and a survey plan planning unit that drafts a leakage investigation plan that determines the execution order of the leakage investigation in the plurality of areas based on the predicted leakage amount information. ,
The water leakage amount prediction unit generates predicted values of both the expected value of the water leakage amount and the uncertainty of the expected value of the water leakage amount as the predicted water leakage amount information, and the investigation plan planning unit defines the water leakage investigation plan as the It is characterized by making use of the leakage cost calculated based on the predicted values of both the expected value and uncertainty of the leakage amount.

  Advantageous Effects of Invention According to the present invention, there is an effect that it is possible to provide a water leakage investigation plan planning device that makes a cost-effective plan.

  Problems, configurations, and effects other than those described above will become apparent from the following description of embodiments.

It is a block diagram of the water leak investigation plan planning apparatus in a present Example. It is a hardware block diagram of the water leak investigation plan planning apparatus in a present Example. It is a figure which shows the water pipe network and water distribution block which a water leak investigation plan planning apparatus makes object of a water leak investigation plan. It is a figure which shows the table pipe line information of the water distribution pipe recorded on a pipe line information storage part. It is a figure which shows the table pipe line information of the water supply pipe recorded on a pipe line information storage part. It is a figure which shows the table water leak investigation information recorded on an investigation and repair log | history memory | storage part. It is a figure which shows the table pipe line repair information recorded on an investigation and repair log | history memory | storage part. It is a figure which shows the example of the performance transition of the amount of leaks of a water distribution block. It is a figure which shows the prediction transition of the water leak amount by a water leak amount estimation part. It is a figure which shows the table prediction model information recorded on a prediction model memory | storage part. It is a figure which shows the table water leak investigation plan information recorded on a water leak plan memory | storage part. It is a flowchart which shows the process of a prediction model learning part. It is a flowchart which shows the process of a water leak amount estimation part. It is a flowchart which shows the process of an investigation plan planning part. It is a figure which shows the screen display by the screen display part regarding a water leak investigation plan. It is a figure which shows the screen display by the screen display part regarding a water leak amount prediction result.

  Hereinafter, embodiments will be described with reference to the drawings. The same reference numerals are assigned to substantially the same parts, and the description will not be repeated.

  FIG. 1 is a block diagram of a water leakage investigation planning apparatus 101 in the present embodiment.

  The water leakage survey planning device 101 includes a water leakage amount estimation unit 110, a prediction model learning unit 111, a water leakage amount prediction unit 112, a survey plan planning unit 113, a pipeline information storage unit 121, and a survey / repair information storage unit. 122, measurement information storage unit 123, water usage storage unit 124, cost information storage unit 125, prediction model storage unit 131, survey plan storage unit 132, survey terminal IF unit 171 and measurement device IF unit 172, a meter reading terminal IF unit 173, and a screen display unit 174.

  The water leakage survey plan planning system 100 includes a water leakage survey plan planning device 101, a survey terminal 181, a measuring device 182, and a meter reading terminal 183.

  The configuration of the water pipe network, water distribution block, and the like that are targeted by the water leakage investigation plan planning device 101 for the water leakage investigation plan will be described later with reference to FIG.

  The water leakage amount estimation unit 110 receives the measurement information recorded in the measurement information storage unit 123 and the water usage information recorded in the water usage storage unit 124, and inputs each of them in a predetermined cycle (for example, one day). The amount of water leakage in the area is estimated, and the estimated amount of water leakage is transmitted to the prediction model learning unit 111 as water leakage amount information.

  The estimated amount of water leakage uses known techniques such as estimation based on the nighttime minimum flow rate and water balance calculation (estimating the amount of water leakage by subtracting the water consumption and other factors from the integrated value of the water distribution). be able to. The amount of water leakage information output from the water leakage amount estimation unit 110 will be supplemented with the description of FIG.

  The prediction model learning unit 111 is recorded in the leakage amount information in the distribution block area received from the leakage amount estimation unit 110, the pipeline information recorded in the pipeline information storage unit 121, and the investigation / repair information storage unit 122. The prediction model storage unit learns the prediction model that calculates the prediction of both the expected value of the future water leakage amount and its uncertainty in each area, and uses the learned prediction model as prediction model information. 131. Details of the process of the prediction model learning unit 111 will be described later with reference to FIG.

  The water leakage amount prediction unit 112 receives the prediction model information recorded in the prediction model storage unit 131 as input, calculates predictions of both the expected value and the uncertainty of the water leakage amount of each area, and calculates the result as predicted water leakage amount information. Is transmitted to the survey plan drafting unit 113 and the screen display unit 174. Details of the processing of the water leakage amount prediction unit 112 will be described later with reference to FIG.

  The investigation plan planning unit 113 receives the leakage amount prediction information received from the leakage amount prediction unit 112 and the cost information recorded in the cost information storage unit 125 and inputs the expected leakage amount as an evaluation index of the leakage investigation plan. Using the leakage cost calculated based on the prediction of both the value and its uncertainty, the leakage calculation plan is calculated and transmitted to the investigation plan storage unit 132.

  Here, the water leakage investigation plan determines the order of the water leakage investigation in a plurality of areas. Leakage investigation means that an operator investigates whether or not there is a leakage that causes a problem by, for example, an audible investigation on a water pipe. Details of the process of the investigation plan planning unit 113 will be described later with reference to FIG.

  The pipe line information storage unit 121 records pipe line information including extension information of the water pipe network with respect to the water pipe network to be planned by the water leakage investigation planning device 101. Specifically, the pipe line information of the water distribution pipe described later in the description of FIG. 4 and the pipe line information of the water supply pipe described later in the description of FIG. 5 are recorded.

  The investigation / repair information storage unit 122 records investigation / repair information including the execution timing of water leakage investigation and pipeline repair for the water pipe network targeted by the water leakage investigation planning device 101. Specifically, the water leakage investigation information described later in the description of FIG. 6 and the pipeline repair information described later in the description of FIG. 7 are recorded.

  The measurement information storage unit 123 records sensor measurement information in the water pipe network to be planned by the water leakage investigation planning device 101. Sensor measurement data (measured values of pressure and flow rate) of each measurement device in a cycle transmitted by the measurement device 182, for example, 1 minute cycle, is recorded in time series.

  The water usage storage unit 124 records water usage information of the water user in the water pipe network to be planned by the water leakage investigation planning device 101. A water user draws water from a distribution pipe of a water pipe network through a water supply pipe, and a water meter for collecting a charge is installed at the drawing point. The amount of water used measured by the water meter is periodically read by a meter reader or an automatic meter reading system. The water usage amount storage unit 124 records the information on the water usage amount of each water user collected in this way, that is, the information on the water usage amount used by a certain water user for a certain period. .

  The cost information storage unit 125 records cost information including costs required for water leakage investigation and pipe repair in the water pipe network to be planned by the water leakage investigation planning device 101 and a loss due to unit water leakage. .

  Here, the cost of water leakage investigation and pipe repair is to conduct water leakage investigation for each area of the water pipe network using a method such as hearing, and also to repair pipes for water leakage found in the investigation. This is a necessary expense. Moreover, the loss due to the unit amount of water leakage is, for example, the cost of leaving a water leak of 1 cubic meter.

  The prediction model storage unit 131 records the prediction model information output from the prediction model learning unit 111. Specifically, it will be described later with reference to FIG.

  The survey plan storage unit 132 records the survey plan information output by the survey plan planning unit 113. Specifically, it will be described later with reference to FIG.

  The survey terminal IF unit 171 transmits the information received from the survey terminal 181 described later to the survey / repair information storage unit 122, and adds the water leakage survey information and the pipeline repair information recorded in the survey / repair information storage unit 122. Or update.

  The measurement device IF unit 172 transmits information received from the measurement device 182 described later to the measurement information storage unit 123, and adds new measurement information to the measurement information storage unit 123.

  The meter-reading terminal IF unit 173 transmits information received from the meter-reading terminal 183, which will be described later, to the water usage storage unit 124, and adds or updates the information on the water usage recorded in the water usage storage unit 124.

  The screen display unit 174 operates the leakage investigation plan planning device 101 through the output device such as a display, for example, the leakage amount prediction information received from the leakage amount prediction unit 112 and the leakage investigation plan information recorded in the investigation plan storage unit 132. Present to the person. Specifically, it is presented using tables and graphs in the format described in FIGS.

  Each measuring device 182 is connected to the water leakage investigation planning device 101 via a communication network. A measuring device such as a flow meter or a pressure gauge installed in the water pipe network to be managed transmits the measured sensor data to the measuring device IF unit 172 of the water leakage investigation planning device 101 via the communication network. A specific example of the measuring device will be described later with reference to FIG.

  Each investigation terminal 181 and meter-reading terminal 183 (a portable information terminal such as a PDA) are connected to the water leakage investigation planning device 101 via a communication network or the like.

  In the case of water leakage investigation and pipeline repair, the survey terminal 181 inputs information on the water leakage survey and pipeline repair by the operation of the user of the survey terminal, and transmits the information to the survey terminal IF unit 171 of the water leak survey planning device 101. To do.

  Further, the meter reading terminal 183 inputs the reading value of the water meter by the operation of the user of the meter reading terminal at the time of meter reading for billing, and uses the information as the meter reading terminal IF of the water leakage investigation planning device 101. To the unit 173.

  FIG. 2 is a hardware block diagram of the water leakage investigation plan planning apparatus 101 in the present embodiment. In FIG. 2, the water leakage investigation planning apparatus 101 includes a CPU 201, a memory 202, a media input / output unit 203, a communication control unit 204, an input unit 205, a display unit 206, a peripheral device IF unit 207, a bus. 210.

  The CPU 201 executes a program on the memory 202. The memory 202 temporarily stores programs, tables, and the like. The media input / output unit 203 is an interface with an information recording medium such as an SD card.

  The communication control unit 204 is connected to the network 220 and performs a communication interface with an external device. The input unit 205 is a user interface including a keyboard and a mouse.

  The display unit 206 is the display described with reference to FIG. The peripheral device IF unit 207 is an interface with devices located in the vicinity such as a printer.

  The bus 210 interconnects the CPU 201, the memory 202, the media input / output unit 203, the communication control unit 204, the input unit 205, the display unit 206, and the peripheral device IF unit 207.

  As is clear from the comparison between FIG. 1 and FIG. 2, the water leakage investigation planning apparatus 101 in FIG. 1 is operated by the CPU 201 executing the program.

  FIG. 3 is a diagram showing a water pipe network and a water distribution block that are targeted for a water leakage investigation plan by the water leakage investigation planning apparatus 101. FIG. 3 shows a distribution reservoir 301 for supplying water to the water pipe network and a water distribution pipe network drawn by a solid line. Further, flow meters 310 to 313 and pressure gauges 321 to 324 are illustrated as measuring devices installed in the water pipe network.

  The distribution block refers to an area in which the flow rate is measured by a flow meter for all of the distribution pipes of water flowing into and out of the area that is set as an area for dividing the water pipe network. In the example of FIG. 3, the areas 331 and 332 are water distribution blocks in which a flow meter 311 and flow meters 312 and 313 are installed in an inflow pipe. The water distribution block is also called a DMA (Distributed Metered Area).

  The investigation plan drafting unit 113 of the leakage investigation plan planning apparatus 101 carries out the leakage investigation in any order as a leakage investigation plan in the plurality of areas for the distribution block or an area obtained by further dividing the distribution block. Develop a plan that defines this. Hereinafter, the unit of water leakage survey including the distribution block is called area.

  Here, the water leakage investigation refers to a task for a person to detect water leakage from a water pipe using devices such as a sound stick, a sound hearing device, and a correlation type water leakage detector. A specific example of the water leakage investigation plan will be described later with reference to FIG.

  FIG. 4 is a diagram illustrating the table pipe line information 400 of the water distribution pipes recorded in the pipe line information storage unit 121. The table water pipe information 400 includes water pipe ID information 401, position information 402, extension information 403, diameter information 404, pipe type information 405, laying year information 406, incidental information 407, and area information 408 in a row. In the table water pipe information 400, all information is recorded as much as possible with one water pipe as one line for all water pipes of the water pipe network to be managed by the water leakage investigation planning apparatus 101. In FIG. 4, only one water pipe is displayed as an example.

  The distribution pipe ID information 401 stores an ID that uniquely defines a specific distribution pipe among all the distribution pipes. The position information 402 stores coordinate information that specifies the position where the water distribution pipe is embedded in cooperation with a GIS (Geographic Information System). In the extension information 403 and the diameter information 404, information on the extension and diameter of the water pipe is stored.

  The pipe type information 405 stores information such as a pipe type such as DIP (ductile cast iron pipe) and CIP (cast iron pipe), pipe specifications (presence of corrosion prevention sleeve), and the like. In the installation year information 406, the installation year of the water pipe is recorded. Alternatively, the burial year may be recorded, and the laying year may be calculated therefrom. The accessory information 407 stores information such as the number and position of accessories such as fire hydrants and air valves, and the number of water pipes to be taken out. The area information 408 stores information on the area (water distribution block) to which the water distribution pipe belongs.

  FIG. 5 is a diagram showing table water supply pipe information 500 indicating the pipe line information of the water supply pipe recorded in the pipe line information storage unit 121. The table water pipe information 500 includes water pipe ID information 501, position information 502, connection destination water pipe ID information 503, extension information 504, diameter information 505, pipe type information 506, age information 507, and area information 508. . In the table water supply pipe information 500, all information is recorded as much as possible with one water supply pipe as one line for all water supply pipes of the water pipe network to be managed by the water leakage investigation planning apparatus 101. In FIG. 5, only one water supply pipe is displayed as an example.

  The water supply pipe ID information 501 stores an ID that uniquely defines a specific water supply pipe among all the water supply pipes. The position information 502 stores coordinate information that specifies the position where the water supply pipe is buried in cooperation with the GIS. The connection destination water pipe ID information 503 stores the ID of the water pipe to which the water pipe is connected.

  The extension information 504 and the diameter information 505 store information related to the extension and diameter of the water supply pipe. The pipe type information 506 stores information on pipe types such as PE (polyethylene pipe) and LP (lead pipe). The laying year information 507 records the laying year of the water supply pipe. Alternatively, the burial year may be recorded. The area information 508 stores information on the area (distribution block) to which the water supply pipe belongs.

  FIG. 6 is a diagram showing table leakage investigation information 600 indicating leakage investigation information recorded in the investigation / repair information storage unit 122. The table water leakage survey information 600 includes survey ID information 601, period information 602, and target area information 603 in a row. In the table water leakage investigation information 600, all information is recorded with the water leakage investigation for one area as one line for the water leakage investigation history conducted in all areas managed by the water leakage investigation planning device 101. . In FIG. 6, only two water leak investigation histories are displayed as an example.

  The survey ID information 601 stores an ID that uniquely defines a specific survey history among all survey histories. The period information 602 stores information on a period during which the survey is performed. The target area information 603 stores the ID of the area where the water leakage investigation was performed in the investigation.

  FIG. 7 is a diagram showing table pipeline repair information recorded in the investigation / repair information storage unit 122. The table pipe repair information 700 includes repair ID information 701, type information 702, pipe ID information 703, position information 704, cause information 705, prevention water leakage information 706, date and time information 707, and survey ID information 708 in a row. In the table pipeline repair information 700, all information is recorded with one pipeline repair as one line for the pipeline repair history carried out for the distribution pipe and the water supply pipe managed by the water leakage investigation planning device 101. Has been. In FIG. 7, one pipeline repair history is displayed as an example.

  The repair ID information 701 stores an ID that uniquely defines a specific repair history among all repair histories. The type information 702 and pipe ID information 703 store information indicating whether the repair target is a distribution pipe or a water supply pipe, and ID information of the target pipe. The position information 704 stores information on the position where the repair is performed in cooperation with the GIS. That is, information on which position of the target pipe has been repaired is stored.

  The cause information 705 stores information on the estimated cause of repaired water leakage. For example, information on the cause is stored as aged deterioration, corrosion, overload, unknown, or the like. The prevention water leakage amount information 706 stores information on the result of estimating the amount of water leakage (flow rate) at the time of repairing the water leakage targeted for the repair at the repair site. The date / time information 707 stores information on the date / time when the repair was performed. In the survey ID information 708, the ID information of the water leak survey that found the water leak targeted for the repair is stored. If the target water leak is not a water leak found in the water leak investigation but a water leak discovered by, for example, a ground water leak report, ID information that can identify the reason for the discovery is stored.

  FIG. 8 is a diagram illustrating an example of a transition of the actual amount of water leakage in the water distribution block. In the graph of FIG. 8, the horizontal axis represents time, and the vertical axis represents the amount of water leakage in a specific area (water distribution block). In general, the amount of water leakage in the area increases monotonously except for the period of the water leakage prevention work indicated by the prevention work period 811 and the prevention work period 812 as indicated by the water leakage quantity 801. Here, the water leakage prevention work refers to water leakage investigation and repair of water leakage discovered in the water leakage investigation.

  As described above, the water leakage amount estimation unit 110 estimates the past water leakage amount of each area as shown in FIG. 8 based on the information recorded in each storage unit.

  FIG. 9 is a diagram showing the predicted transition of the water leakage amount by the water leakage amount prediction unit 112. With reference to FIG. 9, the prediction of both the expected value and the uncertainty of the leakage amount of a specific area (water distribution block) by the leakage amount prediction unit 112 will be described. In the graph of FIG. 9, the horizontal axis indicates the elapsed time since the last water leak investigation and repair in the area, and the vertical axis indicates the amount of water leak in the area.

  The prediction of the transition of the leakage amount output by the leakage amount prediction unit 112 is based on the expected leakage amount expected value 901, the predicted leakage amount lower value 903, and the predicted leakage amount corresponding to the elapsed time since the last leakage investigation and repair. The three high-order values 902 are related to three time series. The high value 902 of the predicted water leakage amount and the low value 903 of the predicted water leakage amount are values for expressing the uncertainty of the expected value 901 of the predicted water leakage amount. For example, the high value and the low value can be the upper limit value and the lower limit value of the 95% confidence interval calculated from, for example, the degree of fit of the prediction model to the past water leakage amount transition.

  In general, the occurrence of water leakage is difficult to treat as a decisive event because there are many uncertain factors involved. For this reason, the prediction based only on the expected value of the water leakage amount cannot be taken into account that the error increases and the uncertainty of the prediction varies from area to area.

  In addition to the expected value as described above, the water leakage survey planning apparatus 101 outputs a high value and a low value as the uncertainty of the expected value, so that a more effective amount of water leakage can be obtained when planning an effective water leakage survey plan. Predictions can be made. As a method of expressing the uncertainty of the expected value, in addition to the high value and the low value, for example, the magnitude of deviation from the expected value may be used.

  FIG. 10 is a diagram showing the table prediction model information 1000 recorded in the prediction model storage unit 131. The table prediction model information 1000 includes prediction model ID information 1001, target area information 1002, model formula information 1003, explanatory variable information 1004, coefficient information 1005, and type information 1006 in columns.

  The table prediction model information 1000 records one prediction model as one row for all prediction models output by the prediction model learning unit 111. In FIG. 10, two prediction models are displayed as an example. The table prediction model information 1000 includes at least one prediction model that can be applied to all areas targeted by the water leakage investigation planning apparatus 101 for planning.

  The prediction model ID information 1001 stores an ID that uniquely defines a prediction model. The target area information 1002 stores an ID of an area to which the prediction model can be applied. For a model that can be applied to a plurality of areas, information that can identify the applicable area is stored.

  The model formula information 1003 stores the number of the mathematical formula used by the prediction model for prediction. The explanatory variable information 1004 stores information on explanatory variables of the area to be used when the prediction model is applied to each area. The prediction model learning unit 111 and the water leakage amount prediction unit 112 acquire specific values of the explanatory variables of the area from the pipe line information storage unit 121 and the investigation / repair information storage unit 122 as necessary. Explanation variables will be described later.

  The coefficient information 1005 stores a specific value of the coefficient of the prediction formula indicated by the model formula information 1003. The type information 1006 stores information indicating whether the prediction model corresponds to the general-purpose prediction model or the area-specific prediction model.

  The prediction model learning unit 111 outputs the prediction model information described with reference to FIG. As described in the type information 1006, the prediction model output by the prediction model learning unit 111 is a plurality of areas by substituting the area-specific prediction model applicable to only one specific area and the value in the area. And general-purpose prediction models that can be applied to

For example, the prediction model whose prediction model ID is E4A332 shown in FIG. 10 is an area-specific prediction model. As a specific example of the prediction formula of the area-specific prediction model, the water leakage survey planning apparatus 101 uses, for example, the following prediction formula. Hereinafter, formulas (1a) to (1c) are collectively referred to as formula (1).
L (t) = L0 + k × t (1a)
Lh (t) = L (t) + Dh0 + mh × t (1b)
Ll (t) = L (t) −D10−ml × t (1c)
here,
L (t): Expected value of predicted water leakage [m ³ / h]
L0: Same as above (immediately after repair of water leakage) [m ³ / h]
Lh (t): High value of predicted water leakage [m ³ / h]
Dh0: Same as above (immediately after water leakage repair) [m ³ / h]
Ll (t): Lower value of predicted water leakage [m ³ / h]
D10: Initial value as above (immediately after water leakage repair) [m ³ / h]
k, mh, ml: positive coefficient t: elapsed time [days]
It is.

For example, the prediction model whose prediction model ID is E3AGE shown in FIG. 10 is a general-purpose prediction model. As a specific example of the prediction formula of the general-purpose prediction model, the water leakage survey planning apparatus 101 uses, for example, the following prediction formula. Hereinafter, Expressions (2a) to (2c) are collectively referred to as Expression (2).
L (t, x) = L0 (x) + k (x) × t (2a)
Lh (t, x) = L (t, x) + Dh0 + mh × t (2b)
Ll (t, x) = L (t, x) −D10−ml × t (2c)

ここで、
Ｌ（ｔ，ｘ） ：予測漏水量の期待値［ｍ^３／ｈ］
Ｌ０（ｘ） ：同上の初期値（漏水修理の直後）［ｍ^３／ｈ］
Ｌｈ（ｔ，ｘ）：予測漏水量の高位値［ｍ^３／ｈ］
Ｄｈ０ ：同上の初期値（漏水修理の直後）［ｍ^３／ｈ］
Ｌｌ（ｔ，ｘ）：予測漏水量の低位値［ｍ^３／ｈ］
Ｄｌ０ ：同上の初期値（漏水修理の直後）［ｍ^３／ｈ］
α０，β０，αｓ，βｓ：係数
ｋ（ｘ），ｍｈ，ｍｌ：正の係数
ｓ：説明変数のインデックス
ｘｓ：インデックスｓの説明変数
ｘ：すべての説明変数
ｔ：経過時間［日］
であり、ｓに関してはすべての説明変数のインデックスについてとる。

here,
L (t, x): Expected value of predicted water leakage [m ³ / h]
L0 (x): Initial value same as above (immediately after water leakage repair) [m ³ / h]
Lh (t, x): High value of predicted water leakage [m ³ / h]
Dh0: Same as above (immediately after water leakage repair) [m ³ / h]
Ll (t, x): Lower value of predicted water leakage [m ³ / h]
D10: Initial value as above (immediately after water leakage repair) [m ³ / h]
α0, β0, αs, βs: coefficients k (x), mh, ml: positive coefficients s: explanatory variable index xs: explanatory variable of index s x: all explanatory variables t: elapsed time [days]
And for s, take the indices of all explanatory variables.

  The explanatory variables of the general-purpose prediction model handled by the water leakage survey planning device 101 include the past number of water leakage repairs in the area, and the number of water leakage repairs that were discovered by reporting rather than the water leakage survey (report leakage repair) Number), the number of water supply pipes in the area, the age of the oldest pipes in the area, for example, the number of aging pipes over 30 years, total extension, etc., in the pipeline information storage unit 121, the survey / repair information storage unit 122, etc. Any index that can be calculated from the stored information can be used.

  Even when the same prediction formula (2) is used, if different sets of explanatory variables are used, the water leakage investigation planning apparatus 101 handles them as different prediction models. Note that the general-purpose prediction model recorded in the prediction model storage unit 131 includes not only the model output by the prediction model learning unit 111 but also a general-purpose prediction model arbitrarily added by the user of the water leakage survey planning device 101. May be.

  For example, a general-purpose prediction model learned in an area that is not a plan target of the water leakage survey planning device 101 is stored in the prediction model storage unit 131, and the water leakage amount prediction unit 112 creates the water leakage survey plan. It is good also as utilizing for the prediction of the amount of water leaks of the planning object area of the apparatus 101. FIG. In the arbitrarily added general-purpose prediction model, the model relating to the expected value is fixed, and in the example of the prediction formula (2), only the coefficient relating to uncertainty, that is, Dh0, D10, mh, ml is learned by the prediction model learning unit 111. It is good.

  FIG. 11 is a diagram showing the table water leakage investigation plan information 1100 recorded in the water leakage plan storage unit 132. As mentioned above, the water leakage investigation plan determines the order of the water leakage investigation in multiple areas. For example, in addition to the execution order, it is possible to determine a period during which a water leakage survey is performed in each area during the planning period. However, in some areas, water leakage surveys may not be conducted. Leakage investigation plan information 1100 defines a period for conducting a water leakage investigation in each area, and in which period of the planning periods 1102 to 1104 the water leakage investigation is performed for each area indicated by the area ID information 1101 Is shown.

  For example, in FIG. 11, in the plan period of 3 years, the area 331 is planned to conduct a water leakage survey three times, the survey period 1111, the survey period 1112 and the survey period 1113, and the area 332 is surveyed 112 times and the survey period 1122 twice. Indicates. The water leakage investigation plan information 1100 stores information on the investigation plan as described above for all areas to be planned.

  FIG. 12 is a flowchart showing the process of the prediction model learning unit 111. FIG. 12 shows an operation flow from the time when the prediction model learning unit 111 extracts based on the recorded information relating to each area, further learns to generate a prediction model, and transmits the prediction model to the prediction model storage unit 131. ing. In start step 1200, the prediction model learning unit 111 starts processing.

  In the input information reception step 1201, the prediction model learning unit 111 receives the pipeline information recorded in the pipeline information storage unit 121 and the survey / repair information recorded in the survey / repair information storage unit 122. In addition, the past leakage amount information of each area is received from the leakage amount estimation unit 110.

  In the water distribution district extraction step 1202, the prediction model learning unit 111 extracts one water distribution district (area) as a learning target of the area-specific prediction model. However, the prediction model learning unit 111 extracts only the area where the transition of the past water leakage amount is estimated by the water leakage amount estimation unit 110.

  In the area-specific prediction model learning step 1203, the prediction model learning unit 111 learns the area-specific prediction model for the extracted area.

  For example, in order to learn an area-specific prediction model using the prediction formula (1) as a prediction formula, the prediction model learning unit 111 receives the prevention work period 811 and the prevention work period 812 in FIG. The initial value L0 and the coefficient k for reproducing the actual data of the amount of water leakage 801 in the period are calculated. For this calculation, a known technique such as a least square method can be used.

  In addition, the prediction model learning unit 111 uses, for example, the prediction formula (1a) obtained as described above as a method for determining the initial values Dh0 and D10, and the coefficients mh and ml, so that the actual data is always the high value Lh (t) and the low value. The minimum initial values Dh0 and D10 and the coefficients mh and ml in the range between the values Ll (t) can be determined.

  In determination step 1204, the prediction model learning unit 111 determines whether or not the area-specific prediction model has been learned for all areas where learning is possible. If there is an area that has not yet been learned, the process returns to the water distribution area extraction step 1202. When learning is completed for all areas where learning is possible, the process proceeds to the general-purpose prediction model extraction step 1205.

  In the general prediction model extraction step 1205, the prediction model learning unit 111 extracts one general prediction model to be learned.

  In the area-specific prediction model extraction step 1206, the prediction model learning unit 111 extracts all areas where a set of explanatory variables used by the general-purpose prediction model to be learned can be calculated, and predicts the area by area from these areas. Extract the model.

  The pipeline information storage unit 121 and the survey / repair information storage unit 122 generally record all information about all the pipes due to the characteristics of the water supply business that generally uses pipes buried underground for several decades. I don't mean. Therefore, depending on the area, information necessary for calculating a specific explanatory variable is not recorded, and a specific explanatory variable may not be calculated. Therefore, in the area-specific prediction model extraction step 1206, the prediction model learning unit 111 extracts an area where all of the pairs of explanatory variables to be used can be calculated and only the area-specific prediction model.

  In the general prediction model learning step 1207, the prediction model learning unit 111 learns the extracted general prediction model. In the learning of the coefficient that determines the general-purpose prediction model, learning is performed based on the extracted coefficient of the area-specific prediction model. Hereinafter, the case where Formula (2) is used for the prediction formula of the general-purpose prediction model and Formula (1) is used for the prediction formula of each area-specific prediction model will be described.

  The index of the extracted area and area-specific prediction model is p, the coefficients of the prediction formula (1) of each area-specific prediction model are L0p, kp, Dh0p, D10p, mhp, mlp, and the prediction values are Lp (t), Lhp. (T) and Llp (t). In addition, the value of the explanatory variable of the index s in the area p is written as xsp, and the explanatory variable in the area p is collectively written as xp.

At this time, the prediction model learning unit 111 calculates the coefficients α0, β0, αs, and βs so that the equations (2d) and (2e) well reproduce the coefficients L0p and kp of each area-specific prediction model. For this calculation, a known technique such as a least square method can be used. Further, the prediction model learning unit 111 uses the prediction formula (2a) obtained above, and sets the minimum coefficients Dh0, D10, mh, and ml so as to satisfy the following two formulas for all areas p. Determine.
Lh (t, xp) ≧ Lhp (t)
Ll (t, xp) ≦ Llp (t)
In determination step 1208, the prediction model learning unit 111 determines whether learning has been performed for all the general-purpose prediction models. If there is a general-purpose prediction model that has not yet been learned, the process returns to step general-purpose prediction model extraction 1205. When learning is completed for all the general-purpose prediction models, the process proceeds to an output information transmission step 1209. In the output information transmission step 1209, the prediction model learning unit 111 transmits the learned prediction model information to the prediction model storage unit 131. In end step 1210, prediction model learning section 111 ends the process.

  FIG. 13 is a flowchart showing the processing of the water leakage amount prediction unit 112. In FIG. 13, the leakage amount prediction unit 112 predicts the future leakage amount based on the prediction model, and selects the predicted leakage amount information based on the prediction model with the smallest difference between the high value and the low value. 3 shows an operation flow until transmission to the investigation plan drafting unit 113 and the screen display unit 173. In step start 1300, the water leakage amount prediction unit 112 starts processing.

  In the input information reception step 1301, the water leakage amount prediction unit 112 receives the prediction model information created by the flow of FIG. 12 from the prediction model storage unit 131. In addition, the water leakage amount prediction unit 112 receives information necessary for calculation of explanatory variables for each area from the pipeline information storage unit 121, the survey / repair information storage unit 122, and the like as necessary.

  In the water distribution area extraction step 1302, the water leakage amount prediction unit 112 extracts one water distribution area (area) as a target of water leakage amount prediction. In the prediction model extraction step 1303, the water leakage amount prediction unit 112 extracts all prediction models applicable to the extracted area from the received prediction model information.

  In the prediction step 1304 based on the general-purpose prediction model, the water leakage amount prediction unit 112 predicts the amount of water leakage in the extracted area using each of the general-purpose prediction models among the extracted prediction models. Here, the prediction of water leakage amount using the general-purpose prediction model by the water leakage amount prediction unit 112 is calculated for the area where the necessary explanatory variable values are extracted, and the calculated explanatory variable values are used for the general-purpose prediction model. Is applied to the process of calculating the expected value, the high value, and the low value of the predicted water leakage amount shown in FIG.

  In determination step 1305, the water leakage prediction unit 112 determines whether the extracted prediction model includes an area-specific prediction model applicable to the extracted area. If it is included, the process proceeds to the prediction step 1306 based on the area-specific prediction model.

  In the prediction step 1306 based on the area-specific prediction model, the water leakage amount prediction unit 112 predicts the amount of water leakage in the area extracted using the area-specific prediction model. Here, the prediction of water leakage refers to the process of calculating the expected value, the high value, and the low value of the predicted water leakage shown in FIG. 9 as in the case of the general-purpose prediction model.

  In the prediction result selection step 1307, the water leakage amount prediction unit 112 compares the prediction results of the leakage amount of each extracted prediction model, selects the prediction model with the smallest difference between the high value and the low value, and selects it. Predicted water leakage information based on the prediction model is used as predicted water leakage information for the area.

  In determination step 1308, the water leakage amount prediction unit 112 determines whether or not the water leakage amount prediction processing has been performed for all areas. If there is an area that has not been predicted yet, the process returns to the water distribution area extraction step 1302. When the prediction process is completed for all areas, the process proceeds to the output information transmission step 1309.

  In the output information transmission step 1309, the water leakage amount prediction unit 112 transmits the calculated water leakage amount prediction information to the investigation plan planning unit 113 and the screen display unit 173. In end step 1310, the water leakage amount prediction unit 112 ends the process.

  When there are a plurality of prediction models that can be applied to a specific area in the leakage amount prediction unit 112, the prediction leakage amount information based on the prediction model with the smallest difference between the high value and the low value is output in the prediction 1306 based on the prediction model for each step area. By doing so, it is possible to output a predicted water leakage amount with smaller uncertainty.

  Note that the leakage amount prediction unit 112 may perform processing in consideration of the update when the pipeline is scheduled to be updated during the target period of the leakage investigation plan. For example, if the update of the pipeline affects the calculation result of the explanatory variable of the area, for the period after the update, the prediction by the general prediction model using the calculation result of the explanatory variable using the information after the pipeline update is performed. Can be output.

  FIG. 14 is a flowchart showing the processing of the investigation plan planning unit 113. FIG. 14 shows an operation flow until the investigation plan planning unit 113 transmits the leakage investigation plan information obtained in consideration of the leakage amount investigation and the cost related to the loss due to leakage to the investigation plan storage unit 132. In start step 1400, the investigation plan planning unit 113 starts processing.

  In the input information reception step 1401, the survey planning unit 113 receives the leakage amount prediction information of each area created by the flow of FIG. 13 from the leakage amount prediction unit 112, and also receives the cost from the cost information storage unit 125. Receive information.

  In the optimization problem configuration step 1402, the survey plan planning unit 113 configures a mathematical optimization problem for formulating a water leakage survey plan. As a constraint condition of the mathematical optimization problem that the survey plan planning unit 113 configures, firstly, in any period of the plan target period, the total number of areas where the water leak investigation is performed simultaneously is made equal to or less than the predetermined number of water leak investigation teams. There is a constraint condition. Secondly, there is a constraint that a period for conducting a water leakage investigation in one area is continuously secured only for a period required for a water leakage investigation in the entire area. The former considers labor costs for water leakage investigation, and the latter considers work efficiency.

  Moreover, the objective function which minimizes the mathematical optimization problem which the investigation plan planning part 113 comprises considered the sum of the evaluation value in both the expense of a water leakage investigation, and the uncertainty of the leakage amount prediction of a water leakage expense, for example. It can be an objective function. Hereinafter, a specific example of the mathematical optimization problem configured by the survey planning unit 113 will be described. As a subscript, an index a of the area and an index t representing each month in the planned period (for example, 3 years) are used.

  As a specific example of determining the period for conducting a water leakage survey in each area during the planning period as a water leakage survey plan, the main decision variable takes a value of 1 only in the month t when the water leakage survey in area a starts. Can be a binary variable y_ {a, t} that takes the value 0.

  If a binary variable z_ {a, t} is defined that takes the value 1 only for the month t when the water leakage investigation is conducted in the area a and takes the value 0 otherwise, the relationship between y_ {a, t} and z_ {a, t} The following constraint conditions are obtained.

ただし、
ｌ＿ａ：エリアａの漏水調査に要する月数（正の整数）
である。この制約式により、一つのエリアで漏水調査を実施する期間は、当該エリアの全域に対する漏水調査の完了に要する期間の長さだけ必ず連続して確保される。

However,
l_a: Number of months required for water leakage survey in area a (positive integer)
It is. With this constraint equation, the period for conducting a water leakage survey in one area is always ensured continuously for the length of the period required to complete the water leakage survey for the entire area.

  Restrictions that limit the total number of areas where water leakage surveys are conducted simultaneously are:

ただし、
Ｔ：漏水調査チーム数（正の整数）
と記述できる。

However,
T: Number of water leakage investigation teams (positive integer)
Can be described.

The objective function f that the investigation planning unit 113 minimizes with the mathematical optimization problem is, for example, the sum of the leakage cost and the investigation cost, using the estimated value CS of the investigation cost and the estimated value CW of the leakage cost as evaluation indexes. It can be the total cost.
f = CW + CS
Here, the survey planning unit 113 uses the water leakage survey cost C_a of the area a recorded in the cost information storage unit 125 as the predicted value CS of the survey cost, for example,

と計算する。

And calculate.

  On the other hand, the investigation planning unit 113 performs calculation based on prediction of both the expected value of the water leakage amount and its uncertainty as the predicted value CW of the water leakage cost.

  For example, for the calculation of the predicted water leakage amount in each area, a parameter expressing the uncertainty between the high value and the low value is set and evaluated in relation to the parameter and the water leakage cost. The maximum water leakage cost is calculated when the parameter takes an arbitrary value within a predetermined uncertainty set.

  In particular,

と計算する。ここで、
ｗ：コスト情報記憶部１２５に記録された漏水の単位量あたりの限界費用
Ｌｈ＿ａ（ｔ；｛ｙ＿｛ａ，τ｝｝）：決定変数ｙ＿｛ａ，ｔ｝を定めた場合の月ｔにおけるエリアａの予測漏水量の高位値
Ｌｌ＿ａ（ｔ；｛ｙ＿｛ａ，τ｝｝）：決定変数ｙ＿｛ａ，ｔ｝を定めた場合の月ｔにおけるエリアａの予測漏水量の低位値
である。

And calculate. here,
w: Marginal cost per unit amount of water leakage recorded in the cost information storage unit 125 Lh_a (t; {y_ {a, τ}}): Area in month t when determining variable y_ {a, t} is defined High value of predicted water leakage amount a1 Ll_a (t; {y_ {a, τ}}): This is the low value of the predicted water leakage amount of area a at month t when the decision variable y_ {a, t} is defined.

  Here, δ_a is a parameter expressing the uncertainty in area a, and the parameter uncertainty set uses a vector δ in which the parameters δ_a for all areas are arranged as elements, and the norm of δ is 1 or less. It is a set.

  In addition, the specific calculation method of Lh_a and Ll_a includes the high and low values of the amount of water leaked from the leak amount prediction unit 112, which are determined by the elapsed time after the water leak investigation / repair, and the investigation / repair information storage unit 122. It is obtained from the time of the last water leakage survey and repair prior to the recorded water leakage survey plan target period, and the time of the water leakage survey and repair during the target period of the water leak survey plan determined by the decision variable.

  In the optimization problem solving step 1403, the investigation plan drafting unit 113 performs the optimization problem solving process configured in the step optimization problem configuration 1402, and converts the obtained optimal solution into the water leakage survey plan information. For example, metaheuristics such as a genetic algorithm or a known technique such as a branch and bound method can be applied to the solution processing.

  In the output information transmission step 1404, the survey plan planning unit 113 transmits the calculated water leakage survey plan information to the survey plan storage unit 132. In an end step 1405, the investigation plan planning unit 113 ends the process.

  It supplements about the predicted value CW of a water leak cost. In general, a water leakage investigation plan obtained using a leakage cost that uses only the expected value of the predicted leakage amount as the objective function of the mathematical optimization problem is a cost (mathematical optimization) when the actual leakage amount deviates from the predicted leakage amount. The objective function value in question) is significantly worse than the evaluation at the time of planning.

  The leakage cost using only the expected value is, for example, CWA,

を用いた目的関数である。ここで、
Ｌ＿ａ（ｔ；｛ｙ＿｛ａ，τ｝｝）：決定変数ｙ＿｛ａ，ｔ｝を定めた場合の月ｔにおけるエリアａの予測漏水量の期待値
である。

Is an objective function using. here,
L_a (t; {y_ {a, τ}}): an expected value of the predicted water leakage amount in area a at month t when the decision variable y_ {a, t} is determined.

  Since the prediction of water leakage always includes uncertainties, the cost of applying the water leakage investigation plan using the expected value is almost certainly greater than the objective function value of the optimal solution of the mathematical optimization problem. Therefore, in the predicted value CW described above, the concept of robust optimization is applied to calculate the cost when a typical error occurs in the predicted water leakage amount. By formulating a water leakage investigation plan using these evaluation values, the cost can be reduced when the water leakage investigation plan is actually applied, compared to the case where an evaluation value using only the expected value of the predicted water leakage amount is used. I can expect.

  The predicted value CW of the water leakage cost is not limited to the above. For example, the water leakage cost may be calculated by giving a large weight to the cost for an area with high uncertainty. Further, the configuration of the mathematical optimization problem is not limited to the above. For example, a predicted value of water leakage cost can be used as an objective function for setting a constraint condition or the like where an upper limit of the survey cost is set and minimizing. As the objective function of the mathematical optimization problem, the water leakage cost of Equation (4) may be simply calculated using the water leakage cost of Equation (5).

  FIG. 15 is a diagram illustrating a screen display by the screen display unit 173 regarding the water leakage investigation plan. The water leakage investigation plan display screen 1501 displayed on the display or the like by the screen display unit 173 includes a water distribution block display 1502, a cost display 1503, and a water leakage investigation plan table display 1504.

  In the water distribution block display 1502, the screen display unit 173 displays the water pipe network, the water distribution block, the area, and the like on the map in cooperation with the GIS for the area that is the planning target of the water leakage investigation planning device 101. In this example, a water distribution block 331 and a water distribution block 332 are displayed.

  The screen display unit 173 displays the cost evaluation result of the water leakage investigation plan in the cost display 1503. In the row of the survey cost 1541, for example, the cost of the water leak survey calculated by the equation (3) is displayed. In the row of the loss cost 1542, for example, the sum of the water leakage costs calculated by the equation (4) is displayed. The total cost 1543 row displays the sum of the survey cost and the loss cost.

  In the column of the robust cost 1532, a cost evaluation value considering the uncertainty of prediction is displayed. For example, the evaluation value according to the evaluation formula (4) is displayed for evaluation of the loss cost. On the other hand, in the column of average cost 1533, the cost evaluation result of the water leakage investigation plan when the average (expected value) of prediction is used is displayed. For example, the evaluation value based on the evaluation formula (5) is displayed instead of the evaluation formula (4) for evaluation of the loss cost.

  The screen display unit 173 displays the leakage leakage investigation plan table display 1504 as a leakage leakage investigation plan table as in FIG.

  FIG. 16 is a diagram illustrating a screen display by the screen display unit 173 regarding the leakage amount prediction result. The water leak prediction display screen 1601 displayed on the display or the like by the screen display unit 173 includes a water distribution block display 1502, a water leak prediction table display 1603, and a water leak prediction graph display 1604.

  In the water leak prediction table display 1603, the screen display unit 173 displays the predicted transition of the water leak amount by the water leak amount prediction unit 112 in a table. In the column of area information 1631, the area ID to be predicted is displayed. In the column of the average predicted value 1632, the expected value of the predicted water leakage amount is displayed. In the column of the high predicted value 1633, the high value of the predicted water leakage amount is displayed. In the column of the lower predicted value 1633, the lower value of the predicted water leakage amount is displayed. When the user of the water leakage investigation planning apparatus 101 performs an operation of changing the display target time, the screen display unit 173 changes the display target time to the specified current (predicted time) or future time. .

  In the water leak prediction graph display 1604, the screen display unit 173 displays the predicted transition of the water leak amount by the water leak amount prediction unit 112 in the same manner as in FIG. Note that the vertical line 1650 indicates the time point (current) of the water leakage investigation plan, and the past leakage amount prediction result is displayed on the left side of the vertical line 1650 and the past leakage amount prediction result on the right side of the vertical line 1650. When the user of the water leakage investigation planning apparatus 101 performs an operation of changing the display target area, the screen display unit 173 changes the display to indicate the water leakage amount prediction result in the designated area.

  The areas where the water leakage investigation planning device 101 predicts the amount of water leakage or makes the investigation plan do not have to be all areas of the water pipe network of the application destination. For example, prediction may be performed only for all areas, and survey planning may be performed for only some areas.

  As described above, with the above-described configuration, the water leakage investigation plan planning apparatus 101 can devise a water leakage investigation plan that can make a cost-effective water leakage investigation plan even when there is uncertainty about water leakage under limited resources.

  The above embodiment is intended for the drafting of a survey plan for water leakage, but the present invention is also applicable to the planning of maintenance plans for facilities other than water supplies, such as the planning of a survey plan for gas leakage. can do.

  In addition, this invention is not limited to an above-described Example, Various modifications are included. For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and are not necessarily limited to those having all the configurations described.

  Each of the above-described configurations, functions, processing units, processing means, and the like may be realized by hardware by designing a part or all of them with, for example, an integrated circuit. Each of the above-described configurations, functions, and the like may be realized by software by interpreting and executing a program that realizes each function by the processor. Information such as programs, tables, and files for realizing each function can be stored in a memory, a hard disk, a recording device such as an SSD (Solid State Drive), or a recording medium such as an IC card, an SD card, or a DVD.

  Further, the control lines and information lines indicate what is considered necessary for the explanation, and not all the control lines and information lines on the product are necessarily shown. Actually, it may be considered that almost all the components are connected to each other.

  100: Leakage investigation plan planning system, 101: Leakage investigation plan planning device, 111: Prediction model learning unit, 112: Leakage amount prediction unit, 113: Investigation plan planning unit, 121: Pipeline information storage unit, 122: Investigation / repair Information storage unit 125: Cost information storage unit 131: Prediction model storage unit 132: Survey plan storage unit 173: Screen display unit

Claims (10)

A water leakage investigation plan drafting device for drafting a water leakage investigation plan for a plurality of areas divided into water pipe networks,
A measurement information collection unit that collects measurement information related to the flow rate of water from a measurement device including a flow meter installed in the water pipe network;
A water usage storage unit for storing water usage information in the area;
A leakage amount estimation unit that estimates a leakage amount of water in the area based on the measurement information and the water usage information;
A pipeline information storage unit that accumulates pipeline information including extension information of the water pipe network in the area;
An investigation / repair information storage unit for accumulating investigation / repair information including the timing of water leakage investigation and pipeline repair in the area;
A prediction model learning unit that generates prediction model information for predicting a transition of the leakage amount of the area based on at least one of the leakage amount information, the pipeline information, and the investigation / repair information;
A water leakage amount prediction unit that generates predicted water leakage amount information of the area based on the prediction model information;
A survey plan drafting unit that formulates a leak survey plan that determines the execution order of the leak survey in the plurality of areas based on the predicted leak amount information;
The leakage amount prediction unit
Generating predicted values of both the expected value of the leakage amount and the uncertainty of the expected value of the leakage amount as the predicted leakage amount information;
The investigation planning department
The water leakage investigation plan planning device, wherein the water leakage investigation plan is planned using a water leakage cost calculated on the basis of both the expected value and the uncertainty of the leakage amount. A water leakage investigation planning apparatus according to claim 1,
The leakage amount prediction unit
As the predicted value of the uncertainty of the expected value of the water leakage amount, both predicted values of the high value and low value of the water leakage amount are generated,
The prediction model learning unit
As the prediction model information, a prediction formula for predicting a high value and a low value of the water leakage amount and a coefficient of the prediction formula are generated. A water leakage investigation planning apparatus according to claim 2,
The leakage amount prediction unit
When there are multiple prediction models that can be applied to one area,
A leakage survey planning apparatus characterized by selecting and generating predicted leakage amount information in which a difference between the high value of the leakage amount and the low value of the leakage amount is the smallest among the corresponding prediction models. A water leakage investigation planning apparatus according to claim 3,
The prediction model learning unit
Each prediction model of the prediction model information to be generated is
An area-specific prediction model that can be applied to a specific area, and a general-purpose prediction model that can be applied to a plurality of areas by substituting explanatory variables of the area calculated based on the pipe network information and the investigation / repair information; Either
The coefficient for determining the general-purpose prediction model is learned based on the coefficient of the area-specific prediction model previously determined. A water leakage investigation planning apparatus according to claim 4,
The investigation planning department
Establish a period for conducting water leakage surveys in each area during the planning period,
The survey planning unit has a first restriction condition in which the total number of areas where the water leakage survey is simultaneously performed in each period of the planning target period is equal to or less than a predetermined number of water leakage survey teams;
Develop a water leakage investigation plan that satisfies the second constraint that the period for conducting a water leakage investigation in one area is secured continuously for the period required to complete the water leakage investigation for the entire area. Leakage investigation planning equipment characterized by The water leakage investigation planning apparatus according to claim 5,
The investigation planning department
The water leakage investigation plan planning device, wherein the water leakage investigation plan is designed to minimize the total cost with respect to the total cost composed of the sum of the water leakage cost and the investigation cost. A water leakage investigation planning apparatus according to claim 6,
The investigation planning department
When evaluating the leakage cost, which is the evaluation index,
In the predicted amount of water leakage in each area, set a parameter that represents the uncertainty between the high and low values,
An apparatus for planning a water leakage investigation characterized by evaluating the relationship between the parameter and the water leakage cost. The water leakage investigation planning apparatus according to claim 7,
The prediction model learning unit
As the explanatory variable used for learning the general-purpose prediction model,
Of the number of leaks repaired in the area, the number of reported leaks repaired in the area, the number of water supply pipes in the area, the number of years of laying the oldest pipes in the area, the number of aging pipes, and the total length of aging pipes A water leakage survey planning device characterized by using at least one. A water leakage investigation planning apparatus according to claim 1;
Measuring devices including flowmeters that transmit measurement information to the water leakage investigation planning device,
And a survey terminal for transmitting survey / repair information including the location of the leak found in the leak survey to the leak survey planning device. A water leakage investigation plan drafting method for drafting a water leakage investigation plan for a plurality of areas divided into water pipe networks,
A collection step for collecting measurement information relating to the flow rate of water;
A storage step for storing water usage information in the area;
An estimation step of estimating a water leakage amount in the area based on the measurement information and the water usage information;
A pipeline information storage step for storing pipeline information including extension information of the water pipe network in the area;
Investigation / repair information accumulation step for accumulating investigation / repair information including the time for conducting water leakage investigation and pipeline repair in the area;
Prediction model information generation step for generating prediction model information for predicting a transition of the leakage amount of the area based on at least one of the leakage amount information, the pipeline information, and the investigation / repair information;
A predicted water leak information generation step for generating predicted water leak information of the area based on the prediction model information;
And a planning step for formulating a water leakage investigation plan that determines an execution order of water leakage investigations in the plurality of areas based on the predicted water leakage information,
The predicted water leakage information generation step includes
Generating predicted values of both the expected value of the leakage amount and the uncertainty of the expected value of the leakage amount as the predicted leakage amount information;
The planning step includes
The water leakage investigation plan planning method characterized in that the water leakage investigation plan is drafted using a water leakage cost calculated based on both predicted values and uncertainties of the water leakage amount.

Images