Granted Patent B2

US 12,481,343 · App. 17/822,024 · Granted Nov 25, 2025

Inventors: Md Arif Khan (Pullman, WA); Gregary C. Zweigle (Pullman, WA); Jared Kyle Bestebreur (Pullman, WA)

Assignee: Schweitzer Engineering Laboratories, Inc.

G06F1/3218G01R25/00G06F3/14H02J13/00002H03H17/04H03H2017/009

Abstract

Claims (57)

1 . A method comprising:

receiving input data associated with a number of assets of an electric power delivery system;

identifying an event associated with a first asset of the number of assets based on a portion of the input data;

generating a number of delayed signals based on the input data;

generating a number of detection signals based on a difference between the delayed signals and the input data;

identifying a subset of the number of detection signals that satisfy a threshold;

sorting the subset of the number of detection signals based at least in part on when a frequency of the subset of the number of detection signals decreases, wherein each detection signal of the number of detection signals is associated with a respective timestamp;

determining that more than one of the detection signals are associated with a common timestamp;

sorting at least the more than one of the detection signals by value referenced to a system frequency of the electric power delivery system to determine a lowest value detection signal;

determining an onset of an event associated with the first asset based on the subset of the number of detection signals and the lowest value detection signal;

generating a display overlay for a portion of the input data associated with the event; and

transmitting the display overlay to a display for presentation via the display with the input data.

2 . The method of claim 1 , wherein the display overlay includes a type of the first asset and the event is a power generation loss for the electric power delivery system.

3 . The method of claim 1 , wherein the threshold is based on a variation of a steady-state frequency of the electric power delivery system.

4 . The method of claim 1 , wherein the delayed signals are generated using an infinite impulse response filter.

5 . The method of claim 4 , wherein the infinite impulse response filter comprises a Butterworth low pass filter.

6 . The method of claim 1 , comprising determining a geographic location of the event within the electric power delivery system based at least in part on the onset of the event.

7 . The method of claim 1 , comprising determining a propagation of the event through the electric power delivery system.

8 . The method of claim 1 , comprising sorting event locations by corresponding sample indices of a time associated with the onset from smallest to largest, wherein a faster frequency drop corresponds to a location closer to a source of the event.

9 . A non-transitory computer-readable storage medium, comprising executable instructions that, when executed by a processor, facilitate performance of operations, comprising:

receiving an input data stream comprising a voltage phase angle of an electric power delivery system;

sorting input data of the input data stream based at least in part on when a frequency of the input data decreases based on timestamps associated with the input data;

generating a preprocessing signal based on a derivative of the voltage phase angle;

performing an event detection operation on the preprocessing signal to detect a power generation event associated with the electric power delivery system;

sorting the input data by timestamp;

determining that at least two signals of the input data comprise a common timestamp;

determining a source of the power generation event by sorting the at least two signals by value referenced to a system frequency of the electric power delivery system to determine a lowest value based on determining that the at least two signals of the input data comprise a common timestamp;

determining a propagation of the power generation event through the electric power delivery system;

generating image content comprising information associated with the source and the propagation; and

transmitting the image content to an information display.

10 . The non-transitory computer-readable storage medium of claim 9 , wherein the image content comprises an indication of the power generation event.

11 . The non-transitory computer-readable storage medium of claim 9 , the operations comprising determining a geographic location of an industrial asset associated with the power generation event.

12 . The non-transitory computer-readable storage medium of claim 9 , wherein the input data stream comprises a voltage magnitude of the electric power delivery system.

13 . The non-transitory computer-readable storage medium of claim 9 , wherein the input data stream comprises a voltage magnitude for the event detection operation comprises:

generating a number of delayed signals based on the preprocessing signal;

generating a number of detection signals based on a difference between the preprocessing signal and the number of delayed signals; and

identifying a subset of the number of detection signals that satisfy a threshold.

14 . The non-transitory computer-readable storage medium of claim 13 , wherein the threshold is based on a variation of a steady-state frequency of the electric power delivery system.

15 . The non-transitory computer-readable storage medium of claim 13 , wherein the number of delayed signals are generated using an infinite impulse response filter.

16 . The non-transitory computer-readable storage medium of claim 15 , wherein the infinite impulse response filter comprises a Butterworth low pass filter.

17 . A method comprising:

receiving timestamped input data corresponding to synchrophasor measurements associated with a plurality of assets of an electric power delivery system;

generating processed signals based on derivatives of voltage phase angles of the timestamped input data;

detecting a power generation event based on a first subset of the processed signals;

in response to detecting the power generation event:

determining one or more metrics based on the first subset of the processed signals;

using the one or more metrics to determine an onset time of the power generation event;

sorting the first subset of the processed signals by timestamp to determine a second subset of the processed signals that crossed a threshold frequency earliest after the onset time of the power generation event;

determining that more than one processed signals of the second subset of processed signals correspond to a common timestamp;

based on determining that the more than one processed signals of the first subset of processed signals correspond to the common timestamp, sorting at least the more than one processed signals by value referenced to a system frequency of the electric power delivery system to determine a lowest value processed signal among the second subset of the processed signals;

determining a source of the power generation event based on a location corresponding to the lowest value processed signal among at least the second subset of the processed signals;

displaying a time series of the timestamped input data via a display; and

overlaying the determined source of the power generation event on the time series via the display.

18 . The method of claim 17 , wherein the one or more metrics comprise a sum of relative frequency signals energy or a mean difference energy, or both.

19 . The method of claim 17 , comprising updating a machine learning model based at least in part on the source and propagation of the power generation event.

20 . The method of claim 17 , wherein the threshold frequency is based on a variation of a steady-state frequency of the electric power delivery system.

21 . The method of claim 17 , wherein the power generation event comprises a power generation loss of at least one generator of the electric power delivery system.

Assignments (1)

Continuity (2)

References Cited (101)