Method and system for automatic residual consumption

• US 9,699,525 B2
• Filed: 01/21/2013
• Issued: 07/04/2017
• Est. Priority Date: 01/21/2013
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Baras, Dorit, Kermany, Einat, Naveh, Yehuda
• Primary Examiner(s): Benlagsir, Amine
• Application Number:
  US13/745,876
• Publication Number:
  US 20140203949A1
• Time in Patent Office: 1,625 Days
• Field of Search: 34087003, 34087001
• CPC Class Codes: H04Q 2209/60 for transmitting utility meters data, i.e. transmission of data from the reader of the utility meter
• CPC Class Codes: H04Q 9/00 Arrangements in telecontrol or telemetry systems for selectively calling a substation from a main station, in which substation desired apparatus is selected for applying a control signal thereto or for obtaining measured values therefrom

Abstract

A computer-implemented method performed by a computerized device, comprising: obtaining consumption data comprising readings indicating consumption of a product, the consumption data is monitored by a plurality of metering devices, wherein the metering devices are associated with a plurality of consumption entities, wherein the plurality of consumption entities comprising a consumption unit and one or more sub consumption units of the consumption unit, wherein the product is supplied serially to the one or more sub consumption units via the consumption unit; and calculating residual consumption of the plurality of consumption entities at a point in time, wherein the residual consumption is consumption of the consumption unit and which is not associated with a sub consumption unit, wherein the residual consumption is calculated based on the consumption by the plurality of consumption entities at a plurality of points in time.

Claims

1. A computer-implemented method performed by a computerized device, comprising:

obtaining consumption readings indicating consumption of a product, from a plurality of metering devices, associated with a plurality of consumption entities, wherein the consumption entities comprising a plurality of consumption units and a corresponding consumption unit of the consumption units comprising a plurality of sub-consumption units, wherein the product is supplied serially to each of the sub-consumption units via the corresponding consumption unit of the consumption units;

calculating, by a processor, residual consumption of the corresponding consumption unit of the consumption units at a point in time, wherein said calculating comprises;

determining a consumption level of the corresponding consumption unit at the point in time;

determining for said each of the sub-consumption units, a corresponding consumption level

determining a first reading of the corresponding consumption level of a first sub-consumption unit of the sub-consumption units at the point in time was not measured by any of the plurality of metering devices, wherein said determining for said each of the sub-consumption units the corresponding consumption level comprises aligning consumption data of the sub-consumption units to the point in time, wherein said aligning comprises, for said each sub-consumption unit of the sub-consumption units;

obtaining one or more readings of the corresponding consumption level of said each sub-consumption unit of the sub-consumption units at a second point in time, wherein the obtained one or more readings was measured by a metering device of the plurality of metering devices at the second point in time; and

calculating an estimated consumption level for the first sub-consumption unit at the point in time based on the obtained one or more readings of the corresponding consumption level of said each of the sub-consumption units at the second point in time; and

subtracting from the consumption level of the corresponding consumption unit at the point in time, a total sum of the consumption level of said each of the sub-consumption units that includes the estimated consumption level of the first sub-consumption unit at the point in time;

determining clusters of the consumption units, wherein each cluster of the clusters comprises the plurality of the consumption units, wherein said determining clusters of the consumption units comprises extracting features from the obtained consumption readings indicating consumption of the product and computing distances therebetween, wherein at least one feature of the extracted features is based on the residual consumption of the corresponding consumption unit;

comparing the calculated residual consumption of the corresponding consumption unit at the point in time with an expected residual consumption based on calculated residual consumption of the consumption units in a same cluster as the corresponding consumption unit;

identifying an anomalous consumption if the comparison indicates a substantial difference between the calculated residual consumption of the corresponding consumption unit and the expected residual consumption based on the same cluster as of the corresponding consumption unit;

wherein said identifying the anomalous consumption is based on a first and second consumptions, wherein the first consumption is associated with a first time resolution and the second consumption is associated with a second time resolution;

wherein the first and second time resolutions are different time resolutions selected from the group consisting of;

an hourly resolution, a daily resolution and a weekly resolution; and

wherein said determining clusters of the consumption units further comprises obtaining external features not derived from the obtained consumption readings indicating consumption of the product.

2. The computer-implemented method of claim 1, wherein the sub consumption units comprise at least two sub consumption units.

3. The computer-implemented method of claim 1, wherein the corresponding consumption unit is a building and wherein the sub consumption units are apartments.

4. The computer-implemented method of claim 1, wherein the consumption data is obtainable from the plurality of metering devices by Automatic Metering Reading (AMR).

5. The computer-implemented method of claim 1, further comprising:

removing from the obtained consumption readings indicating consumption of the product, readings obtained by the metering device in a time period during which the metering device did not monitor consumption of the product.

6. The computer-implemented method of claim 1, wherein the product is selected from the group consisting of:

water, electricity, and gas.

7. The computer-implemented method of claim 1, wherein said determining clusters of the consumption units further comprises extracting features of the residual consumption of the corresponding consumption unit in different time resolutions.

8. A system, comprising:

a plurality of metering devices, wherein a computerized device configured for obtaining consumption data comprising one or more readings indicating consumption of a product, from the plurality of metering devices, associated with a plurality of consumption entities, wherein the consumption entities comprising a plurality of consumption units and a corresponding consumption unit of the consumption units comprising a plurality of sub-consumption units, wherein the product is supplied serially to each of the sub-consumption units via the corresponding consumption unit of the consumption units;

a processor configured for calculating residual consumption of the corresponding consumption unit of the consumption units at a point in time, wherein said calculating comprises;

determining a consumption level of the corresponding consumption unit at the point in time;

determining for said each of the sub-consumption units, a corresponding consumption level

determining a first reading of the corresponding consumption level of a first sub-consumption unit of the sub-consumption units at the point in time was not measured by any of the plurality of metering devices, wherein said determining for said each of the sub-consumption units the corresponding consumption level comprises aligning the consumption data of the sub-consumption units to the point in time, wherein said aligning comprises, for said each sub-consumption unit of the sub-consumption units;

obtaining the one or more readings of the corresponding consumption level of said each sub-consumption unit of the sub-consumption units at a second point in time, wherein the one or more readings was measured by a metering device of the plurality of metering devices at the second point in time; and

calculating an estimated consumption level for the first sub-consumption unit at the point in time based on the obtained one or more readings of the corresponding consumption level of said each of the sub-consumption units at the second point in time; and

subtracting from the consumption level of the corresponding consumption unit at the point in time, a total sum of the consumption level of said each of the sub-consumption units that includes the estimated consumption level of the first sub-consumption unit at the point in time;

wherein said processor is further configured for determining clusters of the consumption units, wherein each cluster of the clusters comprises the plurality of the consumption units, wherein said determining clusters of the consumption units comprises extracting features from the obtained one or more readings and computing distances therebetween, wherein at least one feature of the extracted features is based on the residual consumption of the corresponding consumption unit;

wherein said processor is further configured for comparing the calculated residual consumption of the corresponding consumption unit at the point in time with an expected residual consumption based on calculated residual consumption of the consumption units in a same cluster as the corresponding consumption unit;

wherein said processor is further configured for identifying an anomalous consumption if the comparison indicates a substantial difference between the calculated residual consumption of the corresponding consumption unit and the expected residual consumption based on the same cluster of the corresponding consumption unit;

an input/output unit through which the processor provides an indication of the anomalous consumption;

wherein said identifying the anomalous consumption is based on a first and second consumptions, wherein the first consumption is associated with a first time resolution and the second consumption is associated with a second time resolution;

wherein the first and second time resolutions are different time resolutions selected from the group consisting of;

an hourly resolution, a daily resolution and a weekly resolution; and

wherein said determining clusters of the consumption units further comprises obtaining external features not derived from the obtained one or more readings.

9. The system of claim 8, wherein the computerized device is configured to obtain the consumption data by Automatic Metering Reading (AMR).

10. The system of claim 8, wherein the processor is further adapted to pre-process the consumption data to exclude a portion of the obtained one or more readings and to introduce estimated readings.