Method for locating a leak in a fluid network

• US 10,242,414 B2
• Filed: 06/12/2012
• Issued: 03/26/2019
• Est. Priority Date: 06/12/2012
• Status: Active Grant

• Current Assignee: TaKaDu Ltd.
• Original Assignee: TaKaDu Ltd.
• Inventors: Scolnicov, Haggai, Armon, Amitai, Linhart, Chaim, Bien, Lilach
• Primary Examiner(s): Harrington, Michael P
• Application Number:
  US13/494,411
• Publication Number:
  US 20130332397A1
• Time in Patent Office: 2,478 Days
• CPC Class Codes: F17D 5/02 Preventing, monitoring, or locating loss
• CPC Class Codes: G01M 3/243 for pipes
• CPC Class Codes: G01M 3/28 for pipes, cables or tubes; for pipe joints or seals; for valves ; for welds
• CPC Class Codes: G05B 13/02 electric
• CPC Class Codes: G06F 11/00 Error detection; Error correction; Monitoring error detection, correction or monitoring in information storage based on relative movement between record carrier and transducer G11B20/18; monitoring, i.e. supervising the progress of recording or reproducing G11B27/36; in static stores G11C29/00
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06Q 50/06 Energy or water supply

Abstract

Methods and systems are described for providing cost effective leak detection in a fluid network. Step testing procedures are represented by decision trees and associated expected costs are calculated. Selection of step testing procedures is optimized for low expected cost. The total expected cost rate for a network configuration may be calculated from the rate of leak occurrence and the optimal expected costs associated therewith. Network configuration changes may be recommended by optimizing for total expected cost rate for a fluid network.

Claims

1. A computer implemented method for providing expected cost for locating a leak in a fluid network, the method comprising:

detecting, by a processing device, a leak condition in a flow monitoring zone by the processing device receiving data from a plurality of sensors within a fluid network;

generating, by the processing device, a network trace of the flow monitoring zone, the network trace including data associated with a plurality of valves and a plurality of pipe segments;

identifying, by the processing device, a plurality of sections of the flow monitoring zone in the network trace, wherein the plurality of sections are determined based upon the locations of one or more valves in the flow monitoring zone;

selecting, by the processing device, a step testing procedure from a plurality of step testing procedures based on the identified sections of the flow monitoring zone, the step testing procedure including the identified sections arranged into a decision tree based on the location of valves in the flow monitoring zone, wherein selecting a step testing procedure comprises the processing device;

identifying one or more leaf sections present within the step testing procedure;

computing a probability prob_loc of a leak being located in the section of the fluid network and computing a cost f_loc of locating the leak in the section of the fluid network for each leaf section of the flow monitoring zone; and

calculating the sum of the products of prob_loc and f_loc for all leaf nodes of the step testing procedure, said sum being the expected cost of locating the leak in the fluid network; and

executing, by the processing device, the step testing procedure based on the expected cost of locating the leak, wherein executing the step testing comprises isolating the identified sections by the processing device remotely shutting a plurality of valves within the flow monitoring zone and determining if a leak is present within the sections based upon data received from a plurality of sensors within a fluid network.

2. The method of claim 1 wherein the step of determining the cost f_loc of locating a leak in a section of the fluid network comprises at least one of:

determining a valve configuration cost f_valves due to closing and reopening valves during the step testing procedure;

determining a shutdown costs f_shut due to shutting down service during the step testing procedure;

determining a loss cost f_loss due to expenses incurred as a result of the leak itself during the step testing procedure; and

determining an inspection cost f_inspect incurred by the inspection of the associated section of the fluid network.

3. The method of claim 1 wherein said probability prob_loc of a leak being located in an associated section of the fluid network is determined by analyzing historical data stored in a database.

4. The method of claim 1 wherein said probability prob_loc of a leak being located in an associated section of the fluid network is determined by analyzing flow data obtained from at least one network fluid monitor.

5. The method of claim 1 wherein the step of executing a step testing procedure comprises performing an optimization algorithm on a set of step testing procedures.

6. The method of claim 5 wherein the step of selecting a step testing procedure further comprises using expected cost as a target function.

7. The method of claim 1 wherein the step of selecting a step testing procedure comprises:

calculating an expected cost value for the plurality of step testing procedures for the fluid network; and

selecting the step testing procedure having the lowest expected cost out of said plurality of step testing procedures.

8. The method of claim 1 further comprising:

obtaining at least one possible leak type associated with the leak condition;

determining a predicted occurrence rate rate_leak for each of said at least one possible leak type;

calculating expected cost T(S)_leak of a step testing procedure for each possible leak type; and

calculating the sum of the products of rate_leak and T(S)_leak for all leaf sections of the decision tree said sum being the total expected cost rate for the network configuration.

9. The method of claim 8 wherein said predicted occurrence rate rate_leak is determined by analyzing historical data.

10. A system for providing expected cost for locating a leak in a fluid network, the system comprising:

a water utility network comprising a plurality of pipes and joints, the water utility network being divided into a plurality of flow monitoring zones;

a plurality of sensors located within the water utility network;

a processing device executing a water network analysis engine, the processing device operative to;

receive sensor data from the plurality of sensors;

detect a leak condition in a flow monitoring zone by the processing device receiving the sensor data;

generate a network trace of the flow monitoring zone including data associated with a plurality of valves and a plurality of pipe segments;

identify a plurality of sections of the flow monitoring zone in the network trace, wherein the plurality of sections are determined based upon the locations of one or more valves in the flow monitoring zone;

select a step testing procedure, from a plurality of step testing procedures, based on the identified sections of the flow monitoring zone, the step testing procedure including the identified sections arranged into a decision tree based on the location of valves in the flow monitoring zone, wherein selecting a step testing procedure comprises;

identify one or more leaf sections present within the step testing procedure;

compute a probability prob_loc of a leak being located in the section of the fluid network and computing a cost f_loc of locating the leak in the section of the fluid network for each leaf section of the flow monitoring zone; and

calculate the sum of the products of prob_loc and f_loc for all leaf nodes of the step testing procedure, said sum being the expected cost of locating the leak in the fluid network; and

execute the step testing procedure based on the expected cost of locating the leak, wherein executing the step testing comprises isolating the identified sections by the processing device remotely shutting a plurality of valves within the flow monitoring zone and determining if a leak is present within the sections based upon data received from a plurality of sensors within a fluid network.

11. The system of claim 10 wherein the water analysis engine determines the cost f_loc of locating a leak in a section of the fluid network by determining at least one of:

a valve configuration cost f_valves due to closing and reopening valves during the step testing procedure;

a shutdown cost f_shut due to shutting down service during the step testing procedure;

a loss cost f_loss due to expenses incurred as a result of the leak itself during the step testing procedure; and

an inspection cost f_inspect incurred by the inspection of the associated section of the fluid network.

12. The system of claim 10 wherein the water analysis engine determines the probability prob_loc of a leak being located in an associated section of the fluid network by analyzing historical data stored in a database.

13. The system of claim 10 wherein the water analysis engine determines said probability prob_loc of a leak being located in an associated section of the fluid network by analyzing flow data obtained from at least one network fluid monitor.

14. The system of claim 10 wherein the water analysis engine executes a step testing procedure by performing an optimization algorithm on a set of step testing procedures.

15. The system of claim 14 wherein the water analysis engine selects a step testing procedure by using expected cost as a target function.

16. The system of claim 10 wherein the water analysis engine selects a step testing procedure by calculating an expected cost value for the plurality of step testing procedures for the fluid network and selecting the step testing procedure having the lowest expected cost out of said plurality of step testing procedures.

17. The system of claim 10 wherein the water analysis engine is further operative to:

obtain at least one possible leak type associated with the leak condition;

determine a predicted occurrence rate rate_leak for each of said at least one possible leak type;

calculate expected cost T(S)_leak of a step testing procedure for each possible leak type; and

calculate the sum of the products of rate_leak and T(S)_leak for all leaf sections of the decision tree said sum being the total expected cost rate for the network configuration.

18. The system of claim 17 wherein said predicted occurrence rate rate_leak is determined by analyzing historical data.