Systems and methods for utilizing machine learning to identify non-technical loss
AssigneeC3.ai, Inc.
InventorsThomas M. Siebel, Edward Y. Abbo, Houman Behzadi, Avid Boustani, Nikhil Krishnan, Kuenley Chiu, Henrik Ohlsson, Louis Poirier, Jeremy Kolter
Life of this patent
Estimating the term…
Abstract
Various embodiments of the present disclosure can include systems, methods, and non-transitory computer readable media configured to select a set of signals relating toa plurality of energy usage conditions. Signal values for the set of signals can be determined. Machine learning can be applied to the signal values to identify energy usage conditions associated with non-technical loss.
Claims24
A method for detecting non-technical energy loss, the method comprising:
- generating, by a processor, a plurality of multidimensional representations of a plurality of energy use conditions based on a plurality of feature values extracted from energy data associated with a plurality of meters;
- performing, by the processor, an unsupervised machine learning clustering process on the plurality of multidimensional representations to determine a plurality of clusters;
- and
- outputting, by the processor to a user interface (UI), information that indicates one or more meters of the plurality of meters that are predicted to be associated with non-technical energy loss, the one or more meters associated with one or more energy use conditions identified based on the plurality of clusters.
A system for detecting non-technical energy loss, the system comprising:
- a memory;
- and
- one or more processors coupled to the memory and configured to:
- generate a plurality of multidimensional representations of a plurality of energy use conditions based on a plurality of feature values extracted from energy data associated with a plurality of meters;
- perform an unsupervised machine learning clustering process on the plurality of multidimensional representations to determine a plurality of clusters;
- and
- output, to a user interface (UI), information that indicates one or more meters of the plurality of meters that are predicted to be associated with non-technical energy loss, the one or more meters associated with one or more energy use conditions identified based on the plurality of clusters.
A non-transitory, computer-readable storage device storing instructions that, when executed by one or more processors, cause the one or more processors to perform operations for detecting non-technical energy loss, the operations comprising:
- generating a plurality of multidimensional representations of a plurality of energy use conditions based on a plurality of feature values extracted from energy data associated with a plurality of meters;
- performing an unsupervised machine learning clustering process on the plurality of multidimensional representations to determine a plurality of clusters;
- and
- outputting, to a user interface (UI), information that indicates one or more meters of the plurality of meters that are predicted to be associated with non-technical energy loss, the one or more meters associated with one or more energy use conditions identified based on the plurality of clusters.
Dependencies are inferred from the claim text.