Matching positive transitions in a time trace disaggregation process

1. A method for identifying an electrical device coupled to a power supply, comprising:

detecting a positive transition in a time trace disaggregation process (TTDP);

creating a new instance j;

capturing a positive transition data block for the instance j;

creating a time trace feature description for at least one of the following: power, reactive power, and current;

matching the instance j to a device type by performing a quick match based on at least one of a plurality of quick match parameters;

responsive to a positive identification based on the quick match, performing a match based on said feature description; and

responsive to positive identification, assigning the instance j to the device type.

2. The method of claim 1 , wherein the positive transition data block comprises samples for at least one of the following: power, reactive power, and current.

3. The method of claim 1 , further comprising determining values of at least one of power, reactive power, and current for samples immediately before the positive transition.

4. The method of claim 1 , wherein the matching further comprises:

performing a CalculateMatchError(j,k) function to determine an error value TERR that is a measure of similarity between the instance j and a device k; and

based on a result of the CalculateMatchError(j,k) function, creating a new device using the instance j or assigning the instance j to the device k.

5. The method of claim 4 , wherein performing the CalculateMatchError(j,k) function comprises determining whether the value of TERR is less than a threshold match error value DEVNIM[j] for the device k.

6. The method of claim 5 , further comprising normalizing TERR by dividing TERR by DEVMIN[k] and then comparing the result to ERRMIN.

7. The method of claim 1 , wherein the plurality of quick match parameters comprises at least one of the following: average power during a predefined number of samples, average reactive power during a predefined number of samples, and average current during a predefined number of samples.

8. The method of claim 7 , wherein the plurality of quick match parameters further comprises at least one of the following: generic device type, service leg, and instance code.

9. The method of claim 1 , wherein the matching further comprises performing a MatchInstanceToGenericDevice (j,k) function, the MatchInstanceToGenericDevice (j,k) comprising testing a MAXPOW parameter to determine whether the MAXPOW parameter is within a range of a MAXPOW maximum limit and a MAXPOW minimum limit.

10. The method of claim 9 , wherein performing the MatchInstanceToGenericDevice (j,k) function further comprises testing a MINPOW parameter to determine whether the MINPOW parameter is within a range of a MINPOW maximum limit and a MINPOW minimum limit.

11. The method of claim 10 , wherein performing the MatchInstanceToGenericDevice (j,k) function further comprises testing a POWERFACTOR1 parameter to determine whether the POWERFACTOR1 parameter is within a range of a POWERFACTOR1 maximum limit and a POWERFACTOR1 minimum limit.

12. The method of claim 11 , wherein performing the MatchInstanceToGenericDevice (j,k) function further comprises testing a PEAKCNT parameter to determine whether the PEAKCNT parameter is within a range of a PEAKCNT maximum limit and a PEAKCNT minimum limit.

13. The method of claim 12 , wherein performing the MatchInstanceToGenericDevice (j,k) function further comprises testing a STEPCNT parameter to determine whether the STEPCNT parameter is within a range of a STEPCNT maximum limit and a STEPCNT minimum limit.

14. The method of claim 1 , wherein the matching further comprises performing a CalulateMatchError(j, k) function, the CalulateMatchError(j, k) function comprising:

sequentially selecting all feature triples jj of the feature description of instance j; and

determining whether each of the feature triples jj are marked as used.

15. The method of claim 14 , wherein performing the CalulateMatchError(j, k) function further comprises:

sequentially selecting all feature triples kk of the feature description of device k; and

determining whether each of the features kk are marked as used.

16. The method of claim 15 , further comprising using a FeatureMatch(jj, kk) function to compare the feature triples jj of instance j to the feature triples kk of device k.

17. The method of claim 16 , further comprising using a MisFeatureMatch(jj) function to compute the error contribution of each unused feature triple jj of instance j.

18. The method of claim 1 , further comprising performing a CLUSTER ANALYSIS process to determine whether a device match threshold DEVMIN[ ] is too large such that two or more different devices are combined into a single device.

19. The method of claim 18 , further comprising:

creating new devices so that there is a separate device for each cluster of instances found by the CLUSTER ANALYSIS process; and

selecting a value for the device match threshold DEVMIN[ ] for each separate device to provide sufficient selectivity so that new instances are properly matched to one of the separate devices.

20. The method of claim 1 , further comprising performing a CONSOLIDATION ANALYSIS process to determine if there are multiple devices that can be consolidated.

21. The method of claim 20 , further comprising:

associating the instances of multiple devices with a single device based on a result of the CONSOLIDATION ANALYSIS process; and

adjusting the single device match threshold DEVMIN[ ] to ensure that new instances are correctly matched to the single device.