Leakage Detection and Leakage Location In Supply Networks

• US 20130231876A1
• Filed: 10/27/2011
• Published: 09/05/2013
• Est. Priority Date: 11/05/2010
• Status: Active Grant

• Current Assignee: Siemens AG
• Original Assignee: Jorg Nieveler
• Inventors: Majewski, Kurt, Hanss, Holger, Nieveler, Jrg, Wehrstedt, Jan Christoph
• Granted Patent: US 10,401,250 B2
• Time in Patent Office: Days
• US Class Current: 702/51
• CPC Class Codes: F17D 5/02 Preventing, monitoring, or locating loss
• CPC Class Codes: G01M 3/00 Investigating fluid-tightness of structures
• CPC Class Codes: G01M 3/26 by measuring rate of loss or gain of fluid, e.g. by pressure-responsive devices, by flow detectors
• CPC Class Codes: G01M 3/2807 for pipes G01M3/2892, G01M3/30 take precedence
• CPC Class Codes: G06F 17/10 Complex mathematical operations function generation by table look-up G06F1/03; evaluation of elementary functions by calculation G06F7/544

Abstract

A method and device are provided for leakage detection and leakage location in an area of a supply network (e.g. water supply, gas supply or district heating network), wherein measurement values of sensors of the supply network are statistically analyzed for the presence of leakages using robust regression methods. The false alarm rate (type 2 error) may be reduced or minimized.

Claims

1. A method for leakage detection and leakage location in an area of a supply network, comprising:

dividing the supply network into areas having a comparable consumption profile, wherein the entire water consumption is measurable for each area;

determining a regression curve of the inflow for areas that are comparable in each case, based on past values associated with an assumed leak-free supply network, using robust regression methods,

determining a region of confidence around the regression curve, wherein measurement values within the region of confidence indicate zero leakage;

measuring a respective minimal inflow for each area within a defined measurement period;

generating time series of measurement values of the minimal inflow for selected areas over a number of measurement periods; and

comparing the time series of the measurement values of the minimal inflow with the regression curve, wherein measurement values lying outside the region of confidence indicate the presence of a leakage in the supply network.

2. The method of claim 1, comprising entering the time series of the measurement values into a system of coordinates, wherein each coordinate axis represents an area, wherein the position of a measurement value relative to a coordinate axis indicates a leakage within the corresponding area.

3. The method of claim 1, wherein the quality of the regression curve is defined by a robust correlation coefficient.

4. The method of claim 1, wherein the areas comprise virtual district meter areas.

5. The method of claim 1, wherein sliding mean values or the integrated flow are used instead of the measurement values of the minimal inflow of the areas.

6. The method of claim 1, wherein the measurement period comprises a period from 02:

00 to 04;

00 hours, a period from 00;

00 to 24;

00 hours, or a period from 06;

00 to 18;

00 hours.

7. The method of claim 1, comprising triggering an alarm in response to detecting a measurement value outside the region of confidence.

8. The method of claim 1, wherein the method is used in an infrastructure network for the transportation of a fluid.

9. The method of claim 8, wherein, the infrastructure network is a water supply network or a gas supply network or a district heating network.

10. A device for leakage detection and leakage location in an area of a supply network, in which the supply network is divided into areas having a comparable consumption profile, wherein the entire water consumption is measurable for each area, said device comprising:

instructions stored in non-transitory computer-readable media and executable to;

determine a regression curve of the inflow for areas that are comparable in each case, based on past values associated with an assumed leak-free supply network, using robust regression methods; and

determine a region of confidence around the regression curve, wherein measurement values within the region of confidence indicate zero leakage;

sensors configured to measure a respective minimal inflow for each area within a defined measurement period; and

instructions stored in non-transitory computer-readable media and executable to;

generate time series of measurement values of the minimal inflow for selected areas over a number of measurement periods;

compare the time series of the measurement values of the minimal inflow with a reference regression curve; and

determine the presence of a leakage in the supply network in response to identifying measurement values lying outside the region of confidence.

11. The device of claim 10, comprising instructions stored in non-transitory computer-readable media and executable to enter the time series of the measurement values into a system of coordinates, wherein each coordinate axis represents an area, wherein the position of a measurement value relative to a coordinate axis indicates a leakage within the corresponding area.

12. The device of claim 10, wherein the quality of the regression curve is defined by a robust correlation coefficient.

13. The device of claim 10, wherein the areas comprise virtual district meter areas.

14. The device of claim 10, wherein sliding mean values or the integrated flow are used instead of the measurement values of the minimal inflow of the areas.

15. The device of claim 10, wherein the measurement period comprises a period from 02:

00 to 04;

00 hours, a period from 00;

00 to 24;

00 hours, or a period from 06;

00 to 18;

00 hours.

16. The device of claim 10, instructions stored in non-transitory computer-readable media and executable to trigger an alarm in response to detecting a measurement value outside the region of confidence.

17. A method for leakage detection and leakage location in an area of a supply network, comprising:

dividing the supply network into areas having a comparable consumption profile, wherein the entire water consumption is measurable for each area;

determining a regression curve of the inflow for areas that are comparable in each case, based on past values associated with an assumed leak-free supply network, using robust regression methods,

determining a region of confidence around the regression curve, wherein measurement values within the region of confidence indicate zero leakage;

determining sliding mean values or an integrated flow for each area within a defined measurement period;

generating time series of measurement values of the sliding mean values or integrated flow for selected areas over a number of measurement periods; and

comparing the time series of the measurement values of the sliding mean values or integrated flow with the regression curve, wherein measurement values lying outside the region of confidence indicate the presence of a leakage in the supply network.