Methods and system for nonintrusive load monitoring

• US 9,817,045 B2
• Filed: 12/13/2011
• Issued: 11/14/2017
• Est. Priority Date: 12/13/2010
• Status: Active Grant

• Current Assignee: Fraunhofer USA Inc. (Fraunhofer-Institut für Nachrichtentechnik HHI)
• Original Assignee: Fraunhofer USA Inc. (Fraunhofer-Institut für Nachrichtentechnik HHI)
• Inventors: Zeifman, Michael
• Primary Examiner(s): Desta, Elias
• Application Number:
  US13/993,935
• Publication Number:
  US 20130338948A1
• Time in Patent Office: 2,163 Days
• CPC Class Codes: G01D 2204/14 Displaying of utility usage with respect to time, e.g. for monitoring evolution of usage or with respect to weather conditions
• CPC Class Codes: G01D 2204/24 Identification of individual loads, e.g. by analysing current/voltage waveforms
• CPC Class Codes: G01D 4/002 Remote reading of utility meters
• CPC Class Codes: G01R 21/133 by using digital technique
• CPC Class Codes: G06Q 50/06 Energy or water supply
• CPC Class Codes: Y02B 70/30 Systems integrating technologies related to power network operation and communication or information technologies for improving the carbon footprint of the management of residential or tertiary loads, i.e. smart grids as climate change mitigation technology in the buildings sector, including also the last stages of power distribution and the control, monitoring or operating management systems at local level
• 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 20/242 Home appliances
• CPC Class Codes: Y04S 20/30 Smart metering, e.g. specially adapted for remote reading

Abstract

Identification and tracking of major electric appliances by using aggregate power data obtained at the main breaker level of a residence or commercial establishment. Step power changes and power surges characterize appliances. These features are identified and the time of use and duration statistics are considered to match an observed sequence of power changes with the appliances being turned on and off. The time-dependent usage of appliances and their power consumption are then reconstructed.

Claims

1. A method for nonintrusive monitoring of electrical energy usage by a plurality of electrical loads at a premise to which electrical power is supplied via a common feed, comprising:

clustering, using at least one processor, tuples of power changes in the electrical energy usage at the premise that are attributable to a same load of the plurality of electrical loads at the premise, wherein the clustering comprises analyzing power changes in the electrical energy usage based on usage statistics for the plurality of electrical loads to identify power changes attributable to the same load;

calculating, using the at least one processor, a degree of similarity between clusters of tuples of power changes based at least in part on times associated with power changes in the clusters;

merging, using the at least one processor, clusters of tuples of power changes having a degree of similarity above a threshold;

updating, using the at least one processor, the usage statistics to estimate times different loads of the plurality of electrical loads are in different states; and

based on the updated usage statistics, processing, using the at least one processor, real-time power-change measurements to develop real-time load state information based on detected power changes in the electrical energy usage.

2. The method of claim 1 further including associating the real-time load state information with identifiable loads of the plurality of electrical loads at the premise.

3. The method of claim 2, wherein associating the real-time load state information with identifiable loads comprises correlating known load actions with the updated usage statistics.

4. The method of claim 1 wherein calculating the degree of similarity between clusters and merging clusters based on the degree of similarity comprises applying a Viterbi algorithm to analyze power changes that may be associated with state transitions of known loads, wherein applying the Viterbi algorithm comprises analyzing power changes based at least in part on transition probabilities between states of the known loads.

10. The method of claim 1, further comprising calculating the usage statistics based for the plurality of electrical loads, for use in the clustering, based on an analysis of the electrical energy usage at the premise.

5. A method for nonintrusive monitoring of electrical energy usage by a plurality of electrical loads at a premise to which electrical power is supplied via a common feed, comprising:

a. via a sensor, providing a sensor output signal indicative of power delivered through the feed over a first time interval;

b. monitoring, using at least one processor, the sensor output signal to detect positive and negative changes of power, power surge amplitudes and durations;

c. clustering, using the at least one processor, negative changes of power, wherein clustering negative changes in power comprises;

clustering, using the at least one processor, tuples of power changes in the electrical energy usage at the premise that are attributable to a same load of the plurality of electrical loads at the premise, wherein the clustering comprises analyzing power changes in the electrical energy usage based on usage statistics for the plurality of electrical loads to identify power changes attributable to the same load;

calculating, using the at least one processor, a degree of similarity between clusters of tuples of power changes based at least in part on times associated with power changes in the clusters; and

merging, using the at least one processor, clusters of tuples of power changes having a degree of similarity above a threshold;

d. matching, using the at least one processor, negative and positive changes of power;

e. estimating, using the at least one processor, times loads were turned on and turned off during a second time interval; and

f. statistically, using the at least one processor, characterizing the operation of at least one of said loads.

6. The method of claim 5, wherein:

the clustering further comprises;

g. sorting clusters according to their mean values;

h. applying a Viterbi algorithm to tuples of clusters; and

i. resolving cluster conflicts to provide a set of clusters disambiguating power changes; and

the method further comprises;

j. estimating appliance states based on the power changes represented by said set of clusters.

7. The method of claim 6, further comprising updating clustering and appliance feature statistics based on the results of steps i and j.

8. The method of claim 7, further comprising, for a second time interval shorter than and later than the first time interval, re-performing steps g-j on clusters created from power changes sensed by the sensor during the second time interval.

9. A system for nonintrusive monitoring of electrical energy usage by a plurality of electrical loads at a premise to which electrical power is supplied via a common feed, comprising:

a. a power sensor configured and arranged to monitor aggregate power consumed at the premise;

b. means for detecting changes in the monitored power; and

c. a processor executing stored program instructions to disambiguate the detected power changes and associate them with individual loads by clustering power changes and, by means of a Viterbi algorithm, matching clusters to identify power changes associated with individual loads, wherein clustering the power changes comprises;

clustering tuples of power changes in the electrical energy usage at the premise that are attributable to a same load of the plurality of electrical loads at the premise, wherein the clustering comprises analyzing power changes in the electrical energy usage based on usage statistics for the plurality of electrical loads to identify power changes attributable to the same load;

calculating a degree of similarity between clusters of tuples of power changes based at least in part on times associated with power changes in the clusters; and

merging clusters of tuples of power changes having a degree of similarity above a threshold.