Detection of electric power system anomalies in streaming measurements

Patent metadata

Field	Value
Publication number	US-2018024203-A1
Application number	US-201615214922-A
Country	US
Kind code	A1
Filing date	Jul 20, 2016
Priority date	Jul 20, 2016
Publication date	Jan 25, 2018
Grant date	—

Abstract

The present disclosure pertains to detection of anomalous conditions in a variety of types of systems. In one embodiment, a system may be configured to identify anomalous conditions in a stream of measurements. The system may include a communications interface configured to receive a stream of measurements. An archive subsystem may maintain a data archive comprising a statistical representation of the stream of measurements. A pre-processing subsystem may divide the stream of measurements into a plurality of data windows. The plurality of data windows may be analyzed by an analysis subsystem configured to generate a plurality of normalized representations based on the data archive. The plurality of normalized representations may be grouped into a plurality of ranges. An anomaly detection subsystem may perform a comparison of the plurality of normalized representations to at least one threshold and may determine that the comparison indicates an anomalous condition.

First claim

What is claimed is:

1 . A system configured to identify an anomaly in a stream of measurements, the system comprising:
a communications interface configured to receive a stream of measurements representing electrical conditions within a portion of an electrical power system; an archive subsystem configured to maintain a data archive comprising a statistical representation of the stream of measurements; a pre-processing subsystem configured to divide the stream of measurements into a plurality of data windows; a frequency conversion subsystem configured to transform each of the plurality of data windows to a plurality of frequency domain representations; an analysis subsystem configured to:
normalize the plurality of frequency domain representations based on the data archive;
group the normalized plurality of frequency domain representations into a plurality of frequency ranges; and
an anomaly detection subsystem configured to:
perform a comparison of the normalized plurality of frequency domain representations to at least one threshold and
to determine that the comparison indicates an anomalous condition;
an action subsystem configured to implement an action based on the anomalous condition.

2 . The system of claim 1 , wherein the action is configured to modify the electric power system.

3 . The system of claim 1 , wherein the action comprises one of: a notification of the anomalous condition, a report comprising an indication of the anomalous condition, and flagging a subset of the plurality of data windows associated with the anomalous condition.

4 . The system of claim 1 , wherein the archive subsystem is further configured to add the normalized plurality of frequency domain representations to the data archive.

5 . The system of claim 1 , wherein the normalized plurality of frequency domain representation comprises a z-ratio of one of the plurality of data windows to a distribution represented in the data archive.

6 . The system of claim 1 , wherein the stream of measurements comprises a stream of synchrophaser data.

7 . The system of claim 1 , further comprising an event consolidation subsystem configured to:
detect a plurality of anomalous data windows from among the plurality of data windows; and associate each of the plurality of anomalous data windows with the anomalous condition.

8 . The system of claim 7 , wherein the event consolidation subsystem is further configured to associate the plurality of anomalous data windows based on one of a temporal proximity, a physical proximity, and a severity of the plurality of anomalous data windows.

9 . The system of claim 1 , wherein the anomaly detection subsystem comprises a finite impulse response filter configured to generate an average of the normalized frequency domain representations and to perform the comparison based on the average.

10 . A method for identifying an anomaly in a stream of measurements, the method comprising:
receiving a stream of measurements representing electrical conditions within a portion of an electrical power system; generating a data archive comprising a statistical representation of the stream of measurements; dividing the stream of measurements into a plurality of data windows; transforming each of the plurality of data windows to a plurality of frequency domain representations; generating a normalized plurality of frequency domain representations based on the data archive; grouping the normalized plurality of frequency domain representations into a plurality of frequency ranges; performing a comparison of the normalized plurality of frequency domain representations to at least one threshold; determining that the comparison indicates an anomalous condition; implementing an action based on the anomalous condition.

11 . The method of claim 10 , further comprising modifying the electrical power system based on the action.

12 . The method of claim 10 , wherein the action comprises one of: notifying an operator of the anomalous condition, generating a report comprising an indication of the anomalous condition, and flagging a subset of the plurality of data windows associated with the anomalous condition.

13 . The method of claim 10 , further comprising:
adding the plurality of normalized frequency domain representations to the data archive.

14 . The method of claim 10 , wherein the normalized frequency domain representation comprises a z-ratio of one of the plurality of data windows to a distribution represented in the data archive.

15 . The method of claim 10 , further comprising:
detecting a plurality of anomalous data windows from among the plurality of data windows; and associating each of the plurality of anomalous data windows with the anomalous condition.

16 . The method of claim 15 , wherein the plurality of anomalous data windows are associated based on one of a temporal proximity, a physical proximity, and a severity of the plurality of anomalous data windows.

17 . The method of claim 10 , wherein the plurality of frequency ranges is selected to correspond to an issue in the electric power system.

18 . The method of claim 10 , further comprising generating an average of the normalized frequency domain representations using a finite impulse response filter; and
wherein the comparison is based on the average.

19 . A system configured to identify an anomaly in a stream of measurements, the system comprising:
a communications interface configured to receive a stream of measurements; an archive subsystem configured to maintain a data archive comprising a statistical representation of the stream of measurements; a pre-processing subsystem configured to divide the stream of measurements into a plurality of data windows; an analysis subsystem configured to:
generate a normalized plurality of representations based on the plurality of data windows and the data archive;
group the normalized plurality of representations into a plurality of ranges; and
an anomaly detection subsystem configured to:
perform a comparison of the plurality of normalized representations to at least one threshold and
to determine that the comparison indicates an anomalous condition;
an action subsystem configured to implement an action based on the anomalous condition.

20 . A system of claim 19 , wherein the stream of measurements represents one of a voltage, a current, a power factor, a temperature, and a vibration.

Assignees

• Schweitzer Engineering Lab Inc

Inventors

• Halladay Matthew Jacob
• Blood Ellery Abner

Classifications

• G06F11/1441 Primary

  Resetting or repowering · CPC title

• H03K5/26 Primary

  the characteristic being duration, interval, position, frequency, or sequence · CPC title

• G06F2201/805

  Real-time · CPC title

• G01R31/40 Primary

  Testing power supplies (testing photovoltaic devices H02S50/10) · CPC title