Determining Severity of a Geomagnetic Disturbance on a Power Grid Using Similarity Measures

• US 20170061208A1
• Filed: 08/26/2015
• Published: 03/02/2017
• Est. Priority Date: 08/26/2015
• Status: Active Grant

• Current Assignee: International Business Machines Corporation, Hydro-Quebec (Province of Québec)
• Original Assignee: International Business Machines Corporation, Hydro-Quebec (Province of Québec)
• Inventors: Basu, Chumki, Beland, Jean, Guillon, Sebastien, Kamwa, Innocent
• Granted Patent: US 9,613,271 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 18/22 Matching criteria, e.g. proximity measures
• CPC Class Codes: G06F 2218/08 Feature extraction
• CPC Class Codes: G06F 2218/12 Classification; Matching
• CPC Class Codes: G06V 2201/10 Recognition assisted with metadata
• CPC Class Codes: H02J 13/00002 characterised by monitoring
• CPC Class Codes: H02J 13/00017 using optical fiber
• CPC Class Codes: H02J 2203/20 Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
• CPC Class Codes: H02J 3/00 Circuit arrangements for ac mains or ac distribution networks
• CPC Class Codes: Y02E 40/70 Smart grids as climate change mitigation technology in the energy generation sector
• CPC Class Codes: Y02E 60/00 Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
• CPC Class Codes: Y04S 10/00 Systems supporting electrical power generation, transmission or distribution
• CPC Class Codes: Y04S 10/22 Flexible AC transmission systems [FACTS] or power factor or reactive power compensating or correcting units
• CPC Class Codes: Y04S 10/30 State monitoring, e.g. fault, temperature monitoring, insulator monitoring, corona discharge
• CPC Class Codes: Y04S 40/124 using wired telecommunication networks or data transmission busses

Abstract

Methods, systems, and computer program products for determining severity of a geomagnetic disturbance on a power grid using similarity measures are provided herein. A method includes extracting multiple features of activity associated with geomagnetically induced currents from multiple heterogeneous data streams; comparing the extracted features to historical data related to geomagnetically induced currents to determine a level of similarity between each of the extracted features and the historical data; assigning a level of severity to one or more of the extracted features having a level of similarity with the historical data above a given threshold; and generating a notification comprising (i) a prediction of activity associated with geomagnetically induced currents based on the extracted features having a level of similarity with the historical data above the given threshold, and (ii) the level of severity assigned thereto.

Claims

1. A method, comprising:

extracting multiple features of activity associated with geomagnetically induced currents from multiple heterogeneous data streams, wherein said extracting is executed by a feature extractor component of a computing device;

comparing the multiple extracted features to historical data related to geomagnetically induced currents to determine a level of similarity between each of the multiple extracted features and one or more items of the historical data, wherein said comparing is executed by the computing device;

assigning a level of severity to one or more of the multiple extracted features having a level of similarity with one or more items of the historical data above a given threshold, wherein said assigning is executed by the computing device; and

generating a notification comprising (i) a prediction of activity associated with geomagnetically induced currents based on the one or more extracted features having a level of similarity with one or more items of the historical data above the given threshold, and (ii) the level of severity assigned to each of the one or more extracted features, wherein said generating is executed by the computing device.

2. The method of claim 1, wherein said extracting comprises:

segmenting the multiple heterogeneous data streams into multiple words based on multiple detected events of different types; and

identifying one or more of the words as a feature based on a comparison of the segmented words to a pre-determined collection of words.

3. The method of claim 1, comprising:

implementing a dictionary-based similarity measure on the multiple extracted features to determine one or more regularities across the multiple heterogeneous data streams.

4. The method of claim 1, wherein the multiple heterogeneous data streams comprise data streams derived from harmonics, phasor measurement unit data, geomagnetic/geo-electric data, satellite data and/or space weather forecasts.

5. The method of claim 1, wherein said extracting comprises:

computing self-similarity of module phase A, phase B, and phase C across the multiple heterogeneous data streams.

6. The method of claim 5, comprising:

detecting a feature of activity uniquely associated with geomagnetically induced currents by noting when the self-similarity exceeds a given threshold.

7. The method of claim 1, wherein said comparing comprises applying one or more filters to the multiple extracted features in connection with a sort-merge algorithm.

8. The method of claim 7, wherein the one or more filters are based on synchronicity in behavior of sensors associated with the multiple heterogeneous data streams.

9. The method of claim 7, wherein the one or more filters are based on a trajectory of a given geomagnetic disturbance.

10. The method of claim 1, wherein said comparing comprises implementing one or more normalized compression distance measures to compute similarity of a multi-set of the multiple extracted features to the historical data.

11. The method of claim 1, wherein said assigning comprises computing a pairwise similarity between the one or more extracted features and the one or more items of historical data.

12. The method of claim 11, comprising:

ranking the one or more items of historical data for each of the one or more extracted features according to the corresponding computed pairwise similarity.

13. The method of claim 12, comprising:

computing an average of the top N ranked items of the one or more items of historical data for each of the one or more extracted features, wherein N is a pre-determined value; and

assigning the average as the level of severity for each of the one or more extracted features.

14. A computer program product, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a computing device to cause the computing device to:

extract multiple features of activity associated with geomagnetically induced currents from multiple heterogeneous data streams;

compare the multiple extracted features to historical data related to geomagnetically induced currents to determine a level of similarity between each of the multiple extracted features and one or more items of the historical data;

assign a level of severity to one or more of the multiple extracted features having a level of similarity with one or more items of the historical data above a given threshold; and

generate a notification comprising (i) a prediction of activity associated with geomagnetically induced currents based on the one or more extracted features having a level of similarity with one or more items of the historical data above the given threshold, and (ii) the level of severity assigned to each of the one or more extracted features.

15. The computer program product of claim 14, wherein the multiple heterogeneous data streams comprise data streams derived from harmonics, phasor measurement unit data, geomagnetic/geo-electric data, satellite data and/or space weather forecasts.

16. The computer program product of claim 14, wherein said assigning comprises:

computing a pairwise similarity between the one or more extracted features and the one or more items of historical data;

ranking the one or more items of historical data for each of the one or more extracted features according to the corresponding computed pairwise similarity;

computing an average of the top N ranked items of the one or more items of historical data for each of the one or more extracted features, wherein N is a pre-determined value; and

assigning the average as the level of severity for each of the one or more extracted features.

17. The computer program product of claim 14, wherein the program instructions executable by the computing device further cause the computing device to:

segment the multiple heterogeneous data streams into multiple words based on multiple detected events of different types; and

identify one or more words as a feature based on a comparison of the segmented words to a pre-determined collection of words.

18. The computer program product of claim 14, wherein the program instructions executable by the computing device further cause the computing device to:

implement a dictionary-based similarity measure on the multiple extracted features to determine one or more regularities across the multiple heterogeneous data streams.

19. A system comprising:

a memory; and

at least one processor coupled to the memory and configured for;

extracting multiple features of activity associated with geomagnetically induced currents from multiple heterogeneous data streams;

comparing the multiple extracted features to historical data related to geomagnetically induced currents to determine a level of similarity between each of the multiple extracted features and one or more items of the historical data;

assigning a level of severity to one or more of the multiple extracted features having a level of similarity with one or more items of the historical data above a given threshold; and

generating a notification comprising (i) a prediction of activity associated with geomagnetically induced currents based on the one or more extracted features having a level of similarity with one or more items of the historical data above the given threshold, and (ii) the level of severity assigned to each of the one or more extracted features.

20. A geomagnetically induced currents notification device comprising:

a feature extractor component, executing on the geomagnetically induced currents notification device, to extract multiple features of activity associated with geomagnetically induced currents from multiple heterogeneous data streams; and

a feature selector and comparator component, coupled to the feature extractor component, and executing on the geomagnetically induced currents notification device, to;

compare the multiple extracted features to historical data related to geomagnetically induced currents to determine a level of similarity between each of the multiple extracted features and one or more items of the historical data;

assign a level of severity to one or more of the multiple extracted features having a level of similarity with one or more items of the historical data above a given threshold; and

generate a notification comprising (i) a prediction of activity associated with geomagnetically induced currents based on the one or more extracted features having a level of similarity with one or more items of the historical data above the given threshold, and (ii) the level of severity assigned to each of the one or more extracted features;

wherein the feature extractor component and the feature selector and comparator component are coupled to (i) a central processing node configured to capture data from multiple heterogeneous data sources, and (ii) a display for presentation of the notification to a user.