System and method for identifying likely geographical locations of anomalies in a water utility network

• US 8,583,386 B2
• Filed: 01/18/2011
• Issued: 11/12/2013
• Est. Priority Date: 01/18/2011
• Status: Expired due to Fees

• Current Assignee: TaKaDu Ltd.
• Original Assignee: TaKaDu Ltd.
• Inventors: Armon, Amitai, Scolnicov, Haggai, Linhart, Chaim, Ziv, Raz
• Primary Examiner(s): Bui, Bryan
• Application Number:
  US13/008,819
• Publication Number:
  US 20120185184A1
• Time in Patent Office: 1,029 Days
• Field of Search: 702/51
• US Class Current: 702/51
• CPC Class Codes: G01D 4/002 Remote reading of utility meters
• CPC Class Codes: G05B 23/024 Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks
• CPC Class Codes: Y02B 90/20 Smart grids as enabling technology in buildings sector smart grids supporting the management or operation of end-user stationary applications in general, or like technologies with no associated climate change mitigation effect Y04S20/00
• CPC Class Codes: Y04S 20/30 Smart metering, e.g. specially adapted for remote reading

Abstract

A computerized method for determining one or more statistically likely geographical locations of an anomaly in a region or zone of a water utility network, the water utility network comprising at least a network of pipes for delivering water to consumers and a plurality of meters positioned within the water utility network including a plurality of meters capturing data regarding the water delivering within the region or zone of the anomaly, the method comprising. The method includes receiving anomaly event data, the anomaly event data representing an indication of an anomaly occurring or having occurred within a region or zone of the water utility network, the anomaly event data being associated with meter data produced by one or more of the meters. The method also includes performing a plurality of tests on the anomaly event data each designed to statistically determine a likely geographical location of the anomaly within the region or zone, the performance of each test producing a result. The results of the plurality of tests to generate scores for the determined likely locations for the anomaly are combined. The one or more of the determined likely locations are presented to a user.

Claims

1. A computerized method for determining one or more statistically likely geographical locations of an anomaly in a region or zone of a water utility network, the water utility network comprising at least a network of pipes for delivering water to consumers and a plurality of meters positioned within the water utility network including a plurality of meters capturing data regarding the water delivering within the region or zone of the anomaly, the method comprising:

receiving anomaly event data, the anomaly event data representing an indication of an anomaly occurring or having occurred within a region or zone of the water utility network, the anomaly event data being associated with meter data produced by one or more of the meters;

performing a plurality of tests on the anomaly event data each designed to statistically determine a likely geographical location of the anomaly within the region or zone, the performance of each test producing a result;

combining the results of the plurality of tests to generate scores for the determined likely locations for the anomaly; and

presenting one or more of the determined likely locations to a user.

2. The computerized method of claim 1, comprising receiving the meter data associated with the received anomaly event data, and wherein performing a plurality of tests comprises performing at least one test on the anomaly event data and the associated meter data.

3. The computerized method of claim 2, wherein receiving the meter data comprises receiving meter data representing flow or pressure of the water in the water utility network.

4. The computerized method of claim 1, comprising receiving secondary data representing additional information other than meter data about the water utility network or conditions affecting consumption of water delivered by the water utility network or affecting network operation, and wherein performing the tests comprises performing at least one test on the anomaly event data and received secondary data.

5. The computerized method of claim 4, wherein receiving secondary data comprises receiving secondary data selected from the group consisting of:

map data representing a geographical map of the water utility network;

historical data representing past meter data for the meter data associated with the anomaly event data;

repair data representing one or more repairs performed on the water utility network; and

external data representing weather or other conditions affecting water consumption in the water utility network or affecting network operation.

6. The computerized method of claim 1, wherein performing a plurality of tests comprises performing two or more of the tests in parallel on the anomaly event data.

7. The computerized method of claim 1, wherein performing a plurality of tests comprises performing at least one of the tests on the anomaly event data and the result of another of the tests.

8. The computerized method of claim 1, wherein receiving anomaly event data comprises receiving a plurality of anomaly event data sets over time representing an indication the same anomaly continuing to occur in the water utility network at different times, and wherein performing a plurality of tests comprises repeatedly performing at least one of the tests, each performance being on each received anomaly event data set.

9. The computerized method of claim 1, wherein receiving anomaly event data comprises retrieving the anomaly event data from a database.

10. The computerized method of claim 1, wherein receiving anomaly event data comprises receiving the anomaly event data over a network.

11. The computerized method of claim 1, comprising receiving the meter data associated with the anomaly event data, the meter data including one or more values of parameters affected by the anomaly, and wherein performing a plurality of tests comprises performing at least one of the tests by comparing the one or more affected values of the meter data across a plurality of the meters and identifying one of the meters as being most affected by the anomaly.

12. The computerized method of claim 11, wherein comparing the one or more affected values of the meter data comprises at least one of:

computing an absolute increase in the one or more affected values in each of the meters relative to a predicted or expected value and comparing the absolute increases across the meters;

computing a relative increase in the one or more affected values in each of the meters relative to a predicted or expected value and comparing the relative increases across the meters; and

computing a statistical change in the one or more affected values in each of the meters relative to a predicted or expected value and comparing the statistical changes across the meters.

13. The computerized method of claim 11, wherein the anomaly event data represents an indication of a leak, and wherein comparing one or more affected values comprises comparing flow or pressure values in the meters.

14. The computerized method of claim 1, comprising receiving the meter data associated with the anomaly event data including one or more values of parameters affected by the anomaly, and receiving prior meter data presenting data provided by the meters from one or more times prior to the occurrence of the anomaly including one or more prior values of the parameters affected by the anomaly, and wherein performing a plurality of tests comprises performing at least one of the tests by:

computing a change in the one or more values of the difference between values of a pair of meters hydraulically connected to one another from one or more prior or expected values of that difference;

determining whether any of the one or more computed changes is statistically significant; and

identifying the section between a pair of hydraulically connected meters as a likely location of the anomaly if the change in one or more values as between such hydraulically connected meters is determined to be statistically significant.

15. The computerized method of claim 14, wherein receiving the meter data and prior meter data comprises receiving pressure values and prior pressure values, respectively, and wherein computing a change comprises computing a pressure drop between one or more pairs of hydraulically connected meters and a prior pressure drop between the one or more pairs of hydraulically connected meters.

16. The computerized method of claim 1, comprising receiving prior anomaly event data representing one or more prior occurrences of the same type of anomaly in the water utility network and a geographical location for each of the one or more prior occurrences, and wherein performing a plurality of tests comprises performing at least one of the tests by comparing the anomaly event data to the prior anomaly event data.

17. The computerized method of claim 16, wherein comparing the anomaly event data to the prior anomaly event data comprises:

statistically comparing parameters in the anomaly event data to corresponding parameters of the prior anomaly event data; and

identifying the geographical location of the prior anomaly event data as a likely location of the anomaly if the parameters of the anomaly event data are statistically significantly similar to the parameters of the prior anomaly data.

18. The computerized method of claim 17, comprising calculating relationships between each of the plurality of prior anomalies by clustering the plurality of prior anomaly event data based on associated locations for each of the plurality of previous anomalies, and wherein identifying a likely location of the anomaly comprises comparing anomaly event data with the clustered prior anomaly data and selecting a cluster closest to the associated anomaly data.

19. The computerized method of claim 1 comprising receiving secondary data representing additional information other than meter data about the water utility network or conditions affecting consumption of water delivered by the water utility network, and wherein performing a plurality of tests comprises performing at least one of the tests by correlating the prior anomaly event data and secondary data and comparing the anomaly event data to the correlated prior anomaly event data and secondary data.

20. The computerized method of claim 19 wherein correlating the prior anomaly event data and secondary data comprises constructing a distribution of geographical or network location data of prior anomalies, and wherein identifying a likely location of the anomaly comprises analyzing the anomaly event data against the distribution of geographical or network location data.

21. The computerized method of claim 19 wherein correlating the prior anomaly event data and secondary data comprises constructing a distribution of network data from a second water utility network;

and wherein identifying a likely location of the anomaly comprises analyzing the anomaly event data against the distribution of network data from the second water utility network.

22. The computerized method of claim 1 wherein performing a plurality of tests comprises performing at least of the tests by:

receiving, from at least one of the plurality of meters affected by the anomaly, a time of detection for the anomaly;

receiving data indicating a speed of propagation along the network for the detected anomaly; and

calculating the likely location of the anomaly by combining the time of detection for the anomaly and the speed of propagation to determine a location where differences between measured times of arrival match expected times of arrival.

23. The computerized method of claim 1 wherein presenting the likely location of the anomaly further comprises eliminating a likely location of the anomaly by comparing characteristics of the anomaly and fixed constraints.

24. The computerized method of claim 23, wherein fixed constraints are selected from the set consisting of:

pipe diameter data, service failure data, and network layout data.

25. The computerized method of claim 1 wherein presenting the likely location of the anomaly with a score assigned to the location of the anomaly comprises further presenting a network route likely to contain the likely location of the anomaly.

26. The computerized method of claim 1 comprising reporting characteristics of the anomaly and data associated with related anomalies detected previously.

27. A computerized system for determining one or more likely locations of an anomaly, the one or more likely locations of the anomaly disposed throughout a water utility network, the water utility network comprising at least a network of pipes for delivering water to consumers and a plurality of meters positioned within the water utility network, the system comprising:

a network information database for receiving anomaly data, the anomaly data representing an indication of an anomaly having occurred within the water utility network, the indication of the anomaly being associated with meter data produced by one or more of the meters;

an analysis engine for performing a plurality of separate tests to statistically determine a likely location of the anomaly by statistically combining results of the plurality of tests to generate scores for each of the likely locations for the anomaly; and

a user interface module for presenting the likely locations and associated scores to a user.

28. The computerized system of claim 27 wherein the analysis engine performs steps to calculate the likely location of the anomaly comprising comparing magnitudes of the data produced by the plurality of meters affected by the anomaly;

and reporting an indication of the likely location of the anomaly to a user via a user interface.

29. The computerized system of claim 27, wherein the user interface module further retrieves the likely location of the anomaly and the score assigned to the location of the anomaly from the anomaly database and reporting the location of the anomaly and the score assigned to the location of the anomaly to a user.

30. The computerized system of claim 27 wherein a test performed by the analysis engine comprises:

receiving data indicating the occurrence of a plurality of previous anomalies;

calculating relationships between each of the plurality of previous anomalies by clustering the plurality of previous anomalies based on common events and associated locations for each of the plurality of previous anomalies; and

predicting the location of the anomaly by comparing associated anomaly data with cluster data and selecting the cluster closest to the associated anomaly data.