GB2555535A

GB2555535A - Water leak occurence position estimation device, system, and method

Info

Publication number: GB2555535A
Application number: GB1716409.6A
Authority: GB
Inventors: Adachi Shingo; Takahashi Shinsuke; Koizumi Kenji; Fujii Kenji; Oguma Motoaki; Takemoto Takeshi
Original assignee: Hitachi Ltd
Current assignee: Hitachi Ltd
Priority date: 2015-04-28
Filing date: 2016-03-22
Publication date: 2018-05-02
Anticipated expiration: 2036-03-22
Also published as: WO2016174958A1; GB201716409D0; JP6471035B2; JP2016205065A; GB2555535B

Abstract

The water leak occurrence position estimation device that is disclosed comprises: a measurement value collection unit that collects a first measurement value from a plurality of sensors arranged in a water distribution pipe network; a measurement value prediction unit that predicts a prediction measurement value in a predetermined water distribution state on the basis of the first measurement value of each sensor; an index value calculation unit that, in response to the reception of event data in which a water leak occurrence is treated as an event, calculates an index value from a second measurement value from each sensor after the water leak occurrence and the prediction measurement value, predicted by the measurement value prediction unit, in the same water distribution state as the water distribution state in which the second measurement value was measured; and a location estimation unit that estimates as the location of the water leak occurrence a region that satisfies the positional relation of the arrangement of the plurality of sensors and a water leak occurrence location based on a magnitude comparison between index values of the plurality of sensors.

Description

(56) Documents Cited:

JP 2014/145603 A JP 2009/192329 A JP 2014/222424 A JP 2010/48058 A (58) Field of Search:

INT CL E03B, G01M (71) Applicant(s):

Hitachi, Ltd.

6-6 Marunouchi 1-chome, Chiyodaku 100-8280, Tokyo, Japan (72) Inventor(s):

Shingo Adachi Shinsuke Takahashi Kenji Koizumi Kenji Fujii Motoaki Oguma Takeshi Takemoto (74) Agent and/or Address for Service:

Mewburn Ellis LLP

City Tower, 40 Basinghall Street, LONDON, Greater London, EC2V 5DE, United Kingdom (54) Title of the Invention: Water leak occurence position estimation device, system, and method Abstract Title: Water leak occurence position estimation device, system, and method (57) The water leak occurrence position estimation device that is disclosed comprises: a measurement value collection unit that collects a first measurement value from a plurality of sensors arranged in a water distribution pipe network; a measurement value prediction unit that predicts a prediction measurement value in a predetermined water distribution state on the basis of the first measurement value of each sensor; an index value calculation unit that, in response to the reception of event data in which a water leak occurrence is treated as an event, calculates an index value from a second measurement value from each sensor after the water leak occurrence and the prediction measurement value, predicted by the measurement value prediction unit, in the same water distribution state as the water distribution state in which the second measurement value was measured; and a location estimation unit that estimates as the location of the water leak occurrence a region that satisfies the positional relation of the arrangement of the plurality of sensors and a water leak occurrence location based on a magnitude comparison between index values of the plurality of sensors.

112 Measurement value prediction unit

113 Index value calculation unit

114 Location estimation unit

121 Measurement value storage unit

122 Sensor arrangement storage unit

123 Location storage unit

131 Measurement value collection unit

132 Location display unit

191 Sensor

1/7

100

101

FIG.1

102 χ-⁷

WATER LEAK OCCURRENCE POSITION ESTIMATION DEVICE m 122

113

EVENT DATA

RECEPTION

UNIT

SENSOR ARRANGEMENT STORAGE UNIT

WATER LEAK OCCURRENCE EVENT DETECTION DEVICE

INDEX VALUE CALCULATION UNIT

I

MEASUREMENT VALUE PREDICTION UNIT

POSITION

ESTIMATION

UNIT

112

114

123

		z--- ->
MEASUREMENT VALUE STORAGE UNIT	^121	POSITION STORAGE UNIT

MEASUREMENT

VALUE

COLLECTION

UNIT

131

132

POSITION

DISPLAY

UNIT

MEASUREMENT VALUE COLLECTION DEVICE

SENSOR

103 •191

2/7

F I G. 2

101

F I G. 3

3/7

FIG.4

PRESSURE 1

4/7

F I G. 5

F I G. 6

661 662

663 664

517

F I G. 7

DMA 740 V

701 ά

702

703 ά

704

SENSOR	INDEX VALUE	TYPE
322	4.2	PRESSURE
323	3.8	PRESSURE

6/7

F I G.

301

7/7

F I G. 9

SPECIFICATION

WATER LEAK OCCURRENCE POSITION ESTIMATION DEVICE, SYSTEM, AND

METHOD

TECHNICAL FIELD [0001]

The present invention relates to an estimation device, system and method of estimating a water leak occurrence position in a water distribution pipe network.

BACKGROUND ART [0002]

Patent Document 1 discloses a computerized method for determining geographical positions of an abnormality, the computerized method including: performing a plurality of tests designed to statistically determine a likely geographical position of the abnormality within the region or zone, the performance of each test producing a result; and combining the results of the plurality of tests to generate scores for the determined likely positions for the abnormality.

CITATION LIST

PATENT DOCUMENT [0003]

Patent Document 1: US 8,583,386 B2

SUMMARY OF THE INVENTION

PROBLEMS TO BE SOLVED BY THE INVENTION [0004]

If there is a change in a water distribution state due to other factors almost simultaneously with a water leak occurrence when the water leak occurrence is considered as an abnormality and its position is determined in the method of

Patent Document 1, the change in the water distribution state due to other factors as well as a change in the water distribution state due to the water leak occurrence influence meter data of a pressure gauge. Therefore, in the method of

Patent Document 1, an erroneous position is presented as a water leak occurrence position because of the influence of the change in the water distribution state due to other factors.

Examples of the change in the water distribution state due to other factors include a reduction in an amount of demand due to an influence of the water leak occurrence.

[0005]

The present invention provides a water leak occurrence position estimation device, system and method capable of estimating a water leak occurrence position even when there is an influence of a change in a water distribution state due to other factors including the change in the amount of demand.

SOLUTIONS TO PROBLEMS [0006]

A water leak occurrence position estimation device is disclosed which includes: a measurement value collection unit that collects a first measurement value from a plurality of sensors arranged in a water distribution pipe network; a measurement value prediction unit that predicts a prediction measurement value in a predetermined water distribution state on the basis of the first measurement value of each sensor; an index value calculation unit that, in response to reception of event data in which a water leak occurrence is treated as an event, calculates an index value from a second measurement value from each sensor after the water leak occurrence and the prediction measurement value, predicted by the measurement value prediction unit, in the same water distribution state as the water distribution state in which the second measurement value was measured; and a position estimation unit that estimates as a position of the water leak occurrence a region that satisfies a positional relationship between an arrangement of the plurality of sensors and a water leak occurrence position based on a magnitude comparison among index values of the plurality of sensors.

EFFECTS OF THE INVENTION [0007]

According to the disclosed water leak occurrence position estimation device, it is possible to estimate the water leak occurrence position even when there is the change in the water distribution state due to other factors.

BRIEF DESCRIPTION OF THE DRAWINGS [0008]

Fig. 1 is a configuration diagram of a water leak occurrence position estimation system.

Fig. 2 is a hardware configuration diagram of a water leak occurrence position estimation device.

Fig. 3 is an example of a configuration of DMA.

Fig. 4 is a diagram illustrating time-series changes in measurement values of a sensor associated with a water leak occurrence .

Fig. 5 is a drawing illustrating an example of predicting a measurement value before the water leak occurrence.

Fig. 6 is a drawing illustrating a geographical change in an index value under the condition that a water distribution flow rate including a boundary water head and a water leak amount is the same.

Fig. 7 is an example of screen display of a water leak occurrence position.

Fig. 8 is an example of a configuration of DMA including a plurality of inflow paths.

Fig. 9 is an example of a configuration of DMA in which an acoustic level sensor is installed.

MODE FOR CARRYING OUT THE INVENTION [0009]

Hereinafter, embodiments are described with reference to drawings .

[First Embodiment] [0010]

Fig. 1 illustrates a configuration of a water leak occurrence position estimation system 100 which estimates a water leak occurrence position based on a measurement value of a pressure sensor in a water distribution pipe network. The water leak occurrence position estimation system 100 includes a water leak occurrence position estimation device 101, a water leak occurrence event detection device 102, and a measurement value collection device 103.

[0011]

The measurement value collection device 103 collects a sensor measurement value from a sensor 191 which measures a state of a target water distribution pipe network, and transmits the collected sensor measurement value to the water leak occurrence position estimation device 101 and the water leak occurrence event detection device 102.

[0012]

The water leak occurrence event detection device 102 receives the sensor measurement value from the measurement value collection device 103, detects an abnormality occurrence including the water leak occurrence based on the received sensor measurement value, and transmits information of the detected abnormality occurrence as event data to the water leak occurrence position estimation device 101. The event data transmitted by the water leak occurrence event detection device 102 include information regarding a type of an event, a

DMA (District Metered Area) in which an abnormality occurs, and a time when the abnormality occurs. Further, the type of the event includes at least the water leak occurrence. The water leak occurrence event detection device 102 uses a technique such as a statistical machine learning in the detection of the abnormality occurrence.

[0013]

The water leak occurrence position estimation device 101 includes processing units of an event data reception unit 111, a measurement value prediction unit 112, an index value calculation unit 113, a position estimation unit 114, a measurement value collection unit 131 and a position display unit 132, and storage units of a measurement value storage unit 121, a sensor arrangement storage unit 122 and a position storage unit 123.

[0014]

The event data reception unit 111 receives the event data including the water leak occurrence event from the water leak occurrence event detection device 102, and outputs the received event data to the measurement value prediction unit

112. When the event data is a water leak occurrence event data, the event data reception unit 111 outputs the water leak occurrence event data to the index value calculation unit 113.

[0015]

The measurement value prediction unit 112 reads a measurement value data out from the measurement value storage unit 121, receives the event data from the event data reception unit 111, predicts the sensor measurement value in the same water distribution state as the water distribution state in which the measurement value data is obtained, and outputs the predicted sensor measurement value (hereinafter, a prediction measurement value) to the index value calculation unit 113. Details of the prediction process and the water distribution state will be described later in description of

Fig. 5.

[0016]

The index value calculation unit 113 reads the measurement value out from the measurement value storage unit

121, receives the prediction measurement value from the measurement value prediction unit 112, and receives the water leak occurrence event data from the event data reception unit

111. The index value calculation unit 113 calculates an index value of a positional relationship between a water leak position corresponding to the water leak occurrence event data and a position of the sensor in the pipe network from the measurement value read out from the measurement value storage unit 121 and the prediction measurement value input from the measurement value prediction unit 112 of each sensor in a water leak occurrence DMA included in the event data, and outputs the calculated index value to the position estimation unit 114. Details of the process of calculating the index value will be described later in the description of Fig. 6.

[0017]

The position estimation unit 114 receives the index value of each sensor from the index value calculation unit 113, reads sensor arrangement information out from the sensor arrangement storage unit 122, estimates as a position of the water leak occurrence a region that satisfies a positional relationship between the water leak occurrence position and the position estimation unit 114 based on a magnitude comparison among index values of the sensors, and stores the estimated water leak occurrence position to the position storage unit 123. Details of the process of the position estimation will be described later in the description of Fig.

6.

[0018]

The measurement value storage unit 121 receives the measurement values of the plurality of sensors installed in the water distribution pipe network to be managed from the measurement value collection unit 131. The stored measurement values are read out from the index value calculation unit 113 and the measurement value prediction unit 112.

[0019]

The sensor arrangement storage unit 122 stores the sensor arrangement information including a position of the sensor installed in the water distribution pipe network to be managed

The stored sensor arrangement information is read out by the position estimation unit 114. The sensor arrangement information includes a type of the sensor such as a pressure sensor, a flow rate sensor, and a sound hearing level sensor, and an installation height value of the sensor and the like, in addition to the position of the sensor.

[0020]

The position storage unit 123 receives the water leak occurrence position as a result of estimating the water leak occurrence position from the position estimation unit 114, and stores the received water leak occurrence position. The stored water leak occurrence position is read out by the position display unit 132.

[0021]

The measurement value collection unit 131 receives measurement values of a plurality of sensors installed in the water distribution pipe network to be managed from the measurement value collection device 103, and stores the measurement values to the measurement value storage unit 121.

[0022]

The position display unit 132 reads out the water leak occurrence position as a result of estimating the water leak occurrence position from the position storage unit 123, and presents the water leak occurrence position to an operator of the water leak occurrence position estimation device 101. For example, the position display unit 132 displays the water leak occurrence position on a window of the display for the operator. As long as the operator may recognize the water leak occurrence position, the water leak occurrence position may be displayed on any place like a smart device such as a smart phone or a tablet owned by the operator. Further, when the water leak occurrence position satisfies a predetermined condition, for example, when a specific region is estimated as the water leak occurrence position, the position display unit

132 may perform a push-type notification made by e-mail, an alarm, or the like.

[0023]

Fig. 2 illustrates a hardware configuration of the water leak occurrence position estimation device 101. The water leak occurrence position estimation device 101 is a dedicated

hardware	or a	computer to	which a	CPU 201,	a memo	ry 202, a
communication	control unit	203, an	input unit 204,	a display
unit 205,	and	a peripheral	device	IF unit	206 are	connected
via a bus	210 .
[0024]
The	CPU	201 executes	a program on the	memory	2 02. The

memory 202 stores programs, tables and the like temporarily.

The communication control unit 203 is connected with a network

220. The input unit 204 is a keyboard, a mouse and the like.

The display unit 205 is the display described in Fig. 1. The peripheral device IF unit 206 is an interface such as a printer .

[0025]

As apparent from a comparison between Fig. 1 and Fig. 2, the water leak occurrence position estimation device 101 of

Fig. 1 is realized by the CPU 201 executing the program of each processing unit. The water leak occurrence position estimation device 101 and the water leak occurrence event detection device 102 may be realized as different programs on one dedicated hardware or computer.

[0026]

Fig. 3 illustrates an example of a configuration of one

DMA in water distribution pipe network to be managed by the water leak position estimation device 101.

occurrence

Generally, the DMA is a part of the water distribution pipe network, and the number of pipes having water inflow and outflow with respect to adjacent pipe network is limited to be small (one in many cases), and flow rates are measured in all inflow/outflow pipes. The water distribution pipe network is configured with many DMAs.

[0027]

Fig. 3 is one example of the DMA, which includes a water distribution pipe 351 and the like. An area 340 indicated by a broken line is a DMA where an inf low/out flow pipe is limited to one inflow pipe from a water distribution reservoir 301 and a flow rate sensor 310 is installed in the inflow pipe. Thus, the area 340 is referred to as a DMA 340. In the DMA 340, a plurality of pressure sensors 320 to 325 are installed.

[0028]

The sensor 191 from which the measurement value collection device 103 collects the measurement values are the flow rate sensor 310 and the pressure sensors 320 to 325 in an example of the DMA 340. The pressure sensor 320 measures a pressure at an inflow point into the DMA 340. Hereinafter, the pressure of the inflow point is referred to as an inflow pressure. As in the case of the pressure sensors 321 to 325, sensors located inside the DMA rather than the inflow point of the DMA are referred to as sensors in the pipe network.

[0029]

Fig. 4 is a diagram illustrating a time-series change (trend) in the measurement value of the sensor with the water leak occurrence. With reference to Fig. 4, an example of the measurement value of the sensor stored in the measurement value storage unit 121 and a start of a water leak occurrence position estimation processing will be described. Fig. 4 shows a water distribution flow rate trend 411, an inflow pressure trend 412, and a pressure 1 trend 413 before and after the water leak occurrence. The water distribution flow rate trend

411 is a time-series change in the measurement value of the flow rate sensor 310. The inflow pressure trend 412 is a time series change of the measurement value of the pressure sensor

320. The pressure 1 trend 413 is a time series change in one measurement value of the pressure sensors 321 to 325.

[0030]

As shown in Fig. 4, the measurement value collection unit

131 collects the measurement value of each sensor at a predetermined cycle, for example, at a cycle of 1 to 5 minutes, and stores the collected measurement value in the measurement value storage unit 121. In the example of Fig. 4, time 461 is a water leak occurrence time. At time 462, the water leak occurrence event detection device 102 detects the water leak occurrence and transmits the water leak occurrence event data to the event data reception unit 111.

[0031]

The water leak occurrence position estimation device 101 starts estimating the water leak occurrence position with starting an execution of the index value calculation unit 113 in response to an input of the water leak occurrence event data to the index value calculation unit 113 after the time

462 when the event data reception unit 111 receives the water leak occurrence event data. It may be said that the water leak occurrence position estimation device 101 starts estimating the water leak occurrence position in response to the reception of the water leak occurrence event data by the event data reception unit 111. Thereafter, the water leak occurrence position estimation device 101 repeats the estimation processing of the leakage occurrence position periodically, for example, at times 463 to 465.

[0032]

Fig. 5 is a diagram illustrating an example of prediction of the measurement value by the measurement value prediction unit 112 before the water leak occurrence. The vertical axis is a measurement value of the water distribution flow rate 411 by the flow rate sensor 310. Further, the horizontal axis is a pressure loss, in other words, a difference between a measurement value of the inflow pressure 412 by the pressure sensor 320 and the measurement value 413 of the pressure sensor .

[0033]

The measurement value prediction unit 112 generates a prediction curve (a broken line 511 in Fig. 5) for predicting a measurement value for each sensor or for each time period between event occurrence times included in each event data. In addition to water leak occurrence, types of the event include valve opening, valve closing, water leak isolation (repair) , and so on.

[0034]

In Fig. 5, an actual value (a white circle mark) 501 of a relationship between the water distribution flow rate 411 and the pressure loss represents the actual value in a time zone between a water leak occurrence time included in certain water leak occurrence event data input by the measurement value prediction unit 112 and an occurrence time of the immediately preceding event (an event other than the water leak occurrence event). In addition, an actual value (a hatched shaded circle)

502 indicates an actual value of the time after the water leak occurrence time. In other words, the actual value 501 is the actual value before the water leak occurrence, and the actual value 502 is the actual value after the water leak occurrence.

[0035]

The measurement value prediction unit 112 estimates a prediction curve 511 indicating a correlation between the actual values by using a plurality of the actual values 501 in a time zone before the water leak occurrence time. In addition to the prediction curve 511, the measurement value prediction unit 112 also estimates a prediction curve error range 521-522 indicating an error level of the correlation. A statistical method such as a least-squares method or a statistical machine learning method is used for the prediction curve 511 and the estimation of the prediction curve error range.

[0036]

Subsequently, the measurement value prediction unit 112 calculates a prediction measurement value in the same water distribution state as the input measurement value. The water distribution state is a water distribution flow rate including at least a boundary water head and a water leak amount. The water head at a certain point is a sum of a height of the water distribution pipe at that point and a value obtained by converting a pressure value inside the water distribution pipe to a height. Thus, here, the boundary water head corresponds to the inflow pressure 412.

[0037]

Assuming that the actual value 502 after the water leak occurrence is a point (a point indicating a relationship between the water distribution flow rate and the pressure loss) calculated from the input measurement value, the measurement value prediction unit 112 calculates a pressure loss value at the same water distribution flow rate as the actual value 502 on the prediction curve 511 (obtains a pressure loss corresponding to the water distribution flow rate of the actual value 502 on the prediction curve 511) .

Further, a value obtained by subtracting the calculated pressure loss value from the inflow pressure 412 corresponding to the input measurement value is calculated, and the value is set as the prediction measurement value. In addition, the measurement value prediction unit 112 similarly calculates an error range of the prediction measurement value using the prediction curve error range. The water distribution flow rate of the actual value 502 includes an increase in the flow rate due to the water leak. In addition, the above-mentioned method of selecting the vertical axis and the horizontal axis is an example, and an arbitrary prediction model may be used as long as the prediction can be performed with the water distribution flow rate including the boundary water head and the water leak amount being the same.

[0038]

The measurement value prediction unit 112 generates a prediction curve for each time period between event occurrence times of the events other than the water leak occurrence, such that the water leak occurrence position estimation device 101 can estimate the water leak occurrence position even if the water leak occurs after a plurality of events other than the water leak occurrence occurs. For example, when the water leak occurs several hours after an opening or closing of a valve in conjunction with construction work, a more probable water leak occurrence position can be estimated by generating the prediction curve from actual values after opening or closing the valve.

[0039]

Fig. 6 is a diagram illustrating a geographical change of the index value under the condition that the water distribution flow rate, which is the water distribution state, including the boundary water head and the water leak amount is the same. A real water distribution pipe network spreads twodimensionally, but in Fig. 6, explanation will be made with a one-dimensional example for the sake of convenience. The horizontal axis of the graph in Fig. 6 is a distance from the water distribution reservoir 301. It is indicated that the left end of the horizontal axis is the water distribution reservoir 301 and the distance from the water distribution reservoir 301 increases as it goes to the right end of the horizontal axis.

[0040]

Outflow amounts 611 and 612 indicate distribution of the outflow amount of demand and water leak after and before the water leak occurrence. Since the water distribution flow rate including the water leak amount is the same, an increase in the outflow amount at a water leak occurrence position 650 causes a decrease in the outflow amount at other positions.

[0041]

The water heads 621 and 622 show the distributions of the water heads after and before the water leak occurrence. The measurement value prediction unit 112 calculates from a pressure sensor a value of the water head 622 before the water leak occurrence.

[0042]

A value (difference) obtained by subtracting the water head 622 before the water leak occurrence from the water head

621 after the water leak occurrence is a water head difference

631. Since the height does not change before and after the water leak occurrence, the water head difference 631 is equal to a decrease in the pressure value due to the water leak occurrence .

[0043]

Due to the change in the outflow amount caused by the water leak, the water leak occurrence position is not a point

where the	water head	difference is negative,	for	example,	the
downstream	side of the point 664. Further,	as	indicated by
20 points from 661 to	664, the closer to	the	water	leak
occurrence	position	the point is, the	larger the	head

difference becomes. The relationship between the distance from the water leak occurrence position and the water head difference is not affected by a decrease in demand downstream of the water leak due to pressure drop caused by the water leak.

[0044]

The index value calculation unit 113 calculates as an index value a decrease of a pressure measurement value at each pressure sensor in the pipe network after the water leak occurrence event from a prediction pressure value as an estimated value of the water head difference.

[0045]

In other words, for each pressure sensor in the pipe network, the index value calculation unit 113 subtracts the prediction pressure value of the pressure sensor input from the measurement value prediction unit 112 from the measurement value of the pressure sensor read from the measurement value storage unit 121, and outputs the difference as the index value to the position estimation unit 114.

[0046]

In addition to calculating the index value, the index value calculation unit 113 receives an error range of the prediction pressure value from the measurement value prediction unit 112, and outputs as an error range of the index value a range in which the positive and negative directions of the error range of the prediction pressure value are inverted to the position estimation unit 114.

[0047]

Further, in addition to calculating the index value based on the latest pressure measurement value, the index value calculation unit 113 may calculate the index value for each of a plurality of times after the water leak occurrence time recorded in the received water leak event data and output the index values for the plurality of times to the position estimation unit 114 in time series. As an example of the plurality of times, it is possible to select times at a predetermined interval (a predetermined cycle) after the water leak occurrence time 462, as indicated by each of the times

462 to 465 in Fig. 4.

[0048]

Based on an arrangement of the plurality of sensors and a magnitude comparison between the index values of the plurality of sensors, the position estimation unit 114 estimates as the position of the water leak occurrence a region which is not located downstream of the pressure sensor having a negative value of the index value and is adjacent to the pressure sensor which has a maximum of the index value (greater that those of other pressure sensors) in the DMA.

[0049]

The position estimation unit 114 uses as an upper-andlower stream relationship in the pipe network a fixed upperand-lower-stream relationship stored in the sensor arrangement storage unit 122. For example, the sensor arrangement storage unit 122 may store areas which are obtained by further dividing the DMA 340 finely and an upper-and-lower stream relationship between each of areas and the sensor, and the position estimation unit 114 may use the upper-and-lower stream relation between the area and the sensor to limit the area to be estimated as the water leak occurrence position.

[0050]

The position estimation unit 114 may estimate the water leakage occurrence position based on a magnitude comparison in consideration of the error range by using the error range of the index value input from the index value calculation unit

113. When performing a determination of the pressure sensor having the negative index value, in other words, the magnitude comparison with the index value of 0 (determining whether the index value is positive or negative) and the magnitude comparison among the index values of the pressure sensors for specifying the pressure sensor having the highest index value, the position estimation unit 114 uses, for example, the magnitude comparison in consideration of the following error range .

[0051]

The magnitude comparison between the pressure sensors by the position estimation unit 114 will be described in an example of comparison between the index values of the two pressure sensors A and B. When an error upper limit value of the index value of the pressure sensor B is smaller than an error lower limit value of the index value of the pressure sensor A the position estimation unit 114 determines that the index value of the pressure sensor A is larger than the index value of the pressure sensor B.

[0052]

When the index values of a plurality of times are input in time series from the index value calculation unit 113, the position estimation unit 114 may estimate as the water leak occurrence position a position in which the regions estimated from the index values of the plurality of times are overlapped [0053]

The position estimation unit 114 may estimate the water leak occurrence position from the index value at each time, but may estimate the water leak occurrence position more precisely based on a change of the index value over a long time by narrowing down to the regions estimated as the water leak occurrence position at a plurality of times.

[0054]

By comparing the water head differences on condition that the index value calculation unit 113 and the position estimation unit 114 have the same water distribution flow rate including the water leak amount, the water leak occurrence position estimation device 101 can specify a relationship between the water leak occurrence position and a sensor position even when there is an influence of change in the amount of demand.

[0055]

Fig. 7 illustrates an example of screen display of the water leak occurrence position. The position display unit 132 displays the water leak occurrence position on a screen. A position display window 701 displayed by the position display unit 132 on a display or the like has a DMA selection box 702, a position display panel 703, and an index value display panel

704 .

[0056]

The position display unit 132 displays the water leak occurrence position related to the DMA included in the water leak occurrence event data received by the water leak occurrence position estimation device 101. However, the position display unit 132 changes the DMA to be displayed on the position display panel 703 by the operator operating the

DMA selection box 702.

[0057]

The position display unit 132 displays the water leak occurrence position estimated by the position estimation unit

114 on the position display panel 703. In Fig. 7, the water leak occurrence position is indicated as the shaded portion of the region 761. The position display unit 132 displays the index value of each sensor calculated by the index value calculation unit 113 on the index value display panel 704. In

Fig. 7, a sensor ID, the index value, and a sensor type are displayed.

[0058]

As described above, the water leak occurrence position estimation device 101 can estimate the water leak occurrence position and present the estimated water leak occurrence position to the operator even if there is an influence of a state change due to another factor including the change in the amount of demand.

[Second Embodiment] [0059]

In the present embodiment, a water leak occurrence position estimation device will be described that estimates the upper-and-lower-stream relationship among adjacent sensors and regions based on a pressure sensor measurement value, a sensor installation height value, and a pipe network connection relationship among adjacent sensors. Most of a configuration and processes of the water leak occurrence position estimation device are the same as those in the first embodiment, so the difference will be mainly described.

[0060]

Fig. 8 illustrates a configuration example of one DMA having a plurality of inflow paths in the water distribution pipe network to be subject to water leak occurrence position estimation. As compared with the DMA of Fig. 3, a water distribution reservoir 802, a flow rate sensor 811, and a pressure sensor 829 are added.

[0061]

In the case of the DMA 340 having a plurality of inflow paths (water distribution reservoirs) as shown in Fig. 8, there is a tendency in which a flow direction in the pipe network changes depending on the water distribution state of the boundary water head or the water distribution flow rate in addition to water leak occurrence and the upper-and-lowerstream relationship among adjacent sensors and regions changes [0062]

Therefore, the position estimation unit 114 according to the present embodiment additionally receives the sensor installation height value and the pipe network connection relationship between adjacent sensors from the sensor arrangement storage unit 122, and estimates the upper-andlower-stream relationship among adjacent sensors and regions from the pressure sensor measurement value, the sensor installation height values, and the pipe network connection relationship between adjacent sensors.

[0063]

As shown in the example of Fig. 6, since the water head decreases from the upper stream to the lower stream, the upper-and-lower-stream relationship between the sensors can be estimated by calculating a water head value of the sensor position from the pressure sensor measurement value.

[0064]

The position estimation unit 114 adds the sensor installation height value to a value obtained by converting the pressure sensor measurement value into the height value, and calculates the head value of each sensor. Subsequently, the position estimation unit 114 determines which one of adjacent sensors connected to the pipe network is the upper stream or the lower stream based on a magnitude comparison among the water head values of the sensors. Based on the upper-and-lower-stream relationship and a pipeline connection relationship of each sensor, the upper-and-lower-stream relationship is also estimated for the region near the sensor.

[0065]

The position estimation unit 114 estimates the water leak occurrence position based on the estimated upper-and-lowerstream relationship. Since such an estimation processing is the same as that in the first embodiment, the description thereof is omitted.

[0066]

By dynamically estimating the upper-and-lower-stream relationship from the sensor measurement value, the water leak occurrence position estimation device 101 can estimate the water leak occurrence position more accurately even if there is a DMA or the like having a plurality of inflow paths.

[Third Embodiment] [0067]

In this embodiment, a water leak occurrence position estimation device that manages a water distribution pipe network in which an acoustic level sensor in addition to the pressure sensor is installed will be described. Most of the configuration and processing of the water leak occurrence position estimating device are the same as those in the first embodiment, so differences from the first embodiment will be mainly described.

[0068]

Fig. 9 illustrates a configuration example of one DMA in which an acoustic level sensor is installed in the water distribution pipe network to be subject to the water leak occurrence position estimation. When compared with the DMA of

Fig. 3, acoustic level sensors 941 to 944 are added.

[0069]

For each acoustic level sensor, the measurement value prediction unit 112 calculates, out of the measurement value data before the water leak occurrence, a measurement value data (an acoustic level) in the same time zone on a day different from the day when a measurement value data after the water leak occurrence is calculated, and outputs it as a prediction measurement value to the index value calculation unit 113 as a measurement value.

[0070]

For each acoustic level sensor, the index value calculation unit 113 outputs a value obtained by subtracting the input prediction measurement value from the input measurement value as the index value to the position estimation unit 114.

[0071]

As in the first embodiment, the position estimation unit

114 estimates, as the position of the water leak occurrence, a region in the vicinity of the acoustic level sensor having the largest index value within a range of the water leak occurrence position estimated from information of the pressure sensor, and stores it in the position storage unit 123.

[0072]

The water leak occurrence position estimation device 101 can estimate the water leak occurrence position with higher accuracy by utilizing the measurement value of the acoustic level sensor in addition to the pressure sensor.

[0073]

Embodiments of the present invention are not limited to the above-described embodiments, and various modifications are included. For example, the above-described embodiments have been described in detail in order to explain the present invention in an easy-to-understand manner, and are not necessarily limited to those having all the configurations described. In addition, it is possible to replace a part of a configuration of an embodiment with a configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of the embodiment. Further, it is possible to add, delete, and substitute other configurations with respect to a part of a configuration of each embodiment.

Further, each of the above-described configurations, functions, processing units, processing means, and the like may be realized with hardware, for example, by designing some or all of them with an integrated circuit or the like. In addition, each of the above-described configurations, functions, and the like may be realized with software by a processor interpreting and executing a program that realizes each function.

Information such as a program, a table, a file, and the like that realizes each function can be stored in a memory or a recording device such as a hard disk and an SSD (Solid State

Drive) , or a recording medium such as an IC card, an SD card, or a DVD.

REFERENCE SIGNS LIST [0074]

100: water Leak occurrence position estimation system,

101: water leak occurrence position estimation device, 102:

water leak occurrence event detection device, 103: measurement value collection device, 111: event data reception unit, 112:

measurement value prediction unit, 113: index value calculation unit, 114: position estimation unit, 131:

measurement value collection unit, 132: position display unit.

Claims

1. A water leak occurrence position estimation device, comprising:

a measurement value collection unit that collects a first measurement value from a plurality of sensors arranged in a water distribution pipe network;

a measurement value prediction unit that predicts a prediction measurement value in a predetermined water distribution state on the basis of the first measurement value of each sensor;

an index value calculation unit that, in response to reception of event data in which a water leak occurrence is treated as an event, calculates an index value from a second measurement value from each sensor after the water leak occurrence and the prediction measurement value, predicted by the measurement value prediction unit, in the same water distribution state as the water distribution state in which the second measurement value was measured; and a position estimation unit that estimates as a position of the water leak occurrence a region that satisfies a positional relationship between an arrangement of the plurality of sensors and a water leak occurrence position based on a magnitude comparison among index values of the plurality of sensors.

2. The water leak occurrence position estimation device according to claim 1, wherein the water distribution state is expressed at least in a boundary water head and a water distribution flow rate.

3. The water leak occurrence position estimation device according to claim 2, wherein the plurality of sensors are a plurality of pressure sensors,

10 wherein the index value calculation unit sets as the index value of each of the pressure sensors a decrease value of a pressure measurement value of each of the pressure sensors after the water leak occurrence from a prediction pressure value which is the prediction measurement value, and

15 wherein the position estimation unit estimates as the position of the water leak occurrence a region in the water distribution pipe network which is not located downstream of a first pressure sensor having a negative value of the index value among the plurality of pressure sensors, the region

20 being adjacent to a second pressure sensor which has the index value greater than those of other pressure sensors among the plurality of pressure sensors.

4. The water leak occurrence position estimation device

25 according to claim 2, wherein the index value calculation unit further calculates an error upper limit value and an error lower limit value of the index value, and wherein the position estimation unit estimates the position of the water leak occurrence on the basis of determining that the index value of the first sensor is greater than the index value of the second sensor when the error upper limit value of the index value of the second sensor is smaller than the error lower limit value of the index value of the first sensor with respect to the first sensor and the second sensor of the plurality of sensors.

5. The water leak occurrence position estimation device according to claim 2, wherein the index value calculation unit calculates the index value at a plurality of times after a time of the water leak occurrence included in the event data of the water leak, and wherein the position estimation unit estimates as the position of the water leak occurrence a region where the plurality of the regions estimated from each of the index values at the plurality of the times are overlapped.

6. The water leak occurrence position estimation device according to claim 3, wherein the position estimation unit estimates an upperand-lower-stream relationship between the pressure sensor and the region, which are adjacent thereto, from the pressure measurement value of the pressure sensor, an installation height value of the pressure sensor and a pipe network connection relationship between the position estimation unit and the pressure sensor adjacent thereto, and estimates the position of the water leak occurrence based on the estimated upper-and-lower-stream relationship .

7. The water leak occurrence position estimation device according to claim 3, wherein an acoustic level sensor is placed in the water distribution pipe network, wherein the index value calculation unit sets as the index value of the acoustic level sensor an increase value of an acoustic level measurement value measured by the acoustic level sensor after the water leak occurrence from a prediction acoustic level value which is the prediction measurement value, and wherein the position estimation unit estimates as the position of the water leak occurrence a region which is adjacent to the acoustic level sensor which has a maximum of the index value among the regions estimated based on the index values of the pressure sensors.

8. A water leak occurrence position estimation system which estimates a position of water leak occurrence in a district of a water distribution pipe network, the system comprising:

a measurement value collection device that collects measurement values of a plurality of sensors arranged in a water distribution pipe network;

a water leak occurrence event detection device that detects the water leak occurrence as an event from the measurement values received from the measurement value

collection device; and a water leak occurrence position estimation device, including a measurement value collection unit that collects a first measurement value from the sensors arranged in the water distribution pipe network;

a measurement value prediction unit that predicts a prediction measurement value in a predetermined water distribution state on the basis of the first measurement value of the sensors,

an index value calculation unit that, in response to the reception of event data of an event of the water leak occurrence, calculates an index value from a second measurement value from the sensors after the water leak

occurrence and the prediction measurement value, predicted by the measurement value prediction unit, in the same water distribution state as the water distribution state in which the second measurement value was measured; and a position estimation unit that estimates as a position of the water leak occurrence a region that satisfies a positional relationship between an arrangement of the plurality of sensors and a water leak occurrence position based on a magnitude comparison between index values of the plurality of sensors.

9. A method of estimating a water leak occurrence position in a water leak occurrence position estimation device which estimates a position of water leak occurrence in a district of a water distribution pipe network, the method of estimating the water leak occurrence position comprising:

collecting a first measurement value from a plurality of sensors arranged in a water distribution pipe network;

predicting a prediction measurement value in a predetermined water distribution state on the basis of the first measurement value of the sensors;

calculating, in response to reception of event data in which a water leak occurrence is treated as an event, an index value from a second measurement value from the sensors after the water leak occurrence and the prediction measurement value predicted in the same water distribution state as the water distribution state in which the second measurement value was measured; and estimating as the position of the water leak occurrence a region that satisfies a positional relationship between an arrangement of the plurality of sensors and a water leak occurrence position based on a magnitude comparison between index values of the plurality of sensors.

10. The method of estimating the water leak occurrence position according to claim 9, wherein the water distribution state is expressed at least in a boundary water head and a water distribution flow rate.

11. The method of estimating the water leak occurrence position according to claim 10, wherein the plurality of sensors are a plurality of pressure sensors, and wherein the water leak occurrence position estimation device sets as the index value of each of the pressure sensors a decrease value of a pressure measurement value of each of the pressure sensors after the water leak occurrence from a prediction pressure value which is the prediction measurement value, and estimates as the position of the water leak occurrence a region in the water distribution pipe network which is not located downstream of a first pressure sensor having a negative value of the index value among the plurality of pressure sensors, the region being adjacent to a second pressure sensor which has the index value greater than those of other pressure sensors among the plurality of pressure sensors .

12. The method of estimating the water leak occurrence position according to claim 10, wherein the water leak occurrence position estimation device further calculates an error upper limit value and an error lower limit value of the index value, and estimates the position of the water leak occurrence on the basis of determining that the index value of the first sensor is greater than the index value of the second sensor when the error upper limit value of the index value of the second sensor is smaller than the error lower limit value of the index value of the first sensor with respect to the first sensor and the second sensor of the plurality of sensors.

13. The method of estimating the water leak occurrence position according to claim 10, wherein the water leak occurrence position estimation device calculates the index value at a plurality of times after a time of the water leak occurrence included in the event data of the water leak, and estimates as the position of the water leak occurrence a region where the plurality of the regions estimated from each of the index values at the plurality of the times are overlapped.

14. The method of estimating the water leak occurrence position according to claim 11, wherein the water leak occurrence position estimation device estimates an upper-and-lower-stream relationship between the pressure sensor and the region, which are adjacent thereto, from the pressure measurement value of the pressure sensor, an installation height value of the pressure sensor and a pipe network connection relationship between the water leak occurrence position estimation device and the pressure sensor adjacent thereto, and estimates the position of the water leak occurrence based on the estimated upper-and-lowerstream relationship.

15. The method of estimating the water leak occurrence position according to claim 11, wherein an acoustic level sensor is placed in the water distribution pipe network, and wherein the water leak occurrence position estimation device

5 sets as the index value of the acoustic level sensor an increase value of an acoustic level measurement value measured by the acoustic level sensor after the water leak occurrence from a prediction acoustic level value which is the prediction measurement value, and

10 estimates as the position of the water leak occurrence a region which is adjacent to the acoustic level sensor which has a maximum of the index value among the regions estimated based on the index values of the pressure sensors .