NATURAL LANGUAGE NAVIGATION FOR POWER MONITORING SYSTEMS

• US 20110025516A1
• Filed: 07/29/2009
• Published: 02/03/2011
• Est. Priority Date: 07/29/2009
• Status: Active Grant

• Current Assignee: Schneider Electric USA, Inc. (Schneider Electric SE)
• Original Assignee: Square D Company (Schneider Electric SE)
• Inventors: Thompson, Gregory Allen, Johnson, Jeffrey Wayne, Wall, Theresa Krivanek, Stevenson, Thomas Stewart
• Granted Patent: US 8,253,580 B2
• Time in Patent Office: Days
• US Class Current: 340/657
• CPC Class Codes: G06F 16/24522 Translation of natural language queries to structured queries
• CPC Class Codes: G06F 40/205 Parsing
• CPC Class Codes: G06F 40/274 Converting codes to words; Guess-ahead of partial word inputs

Abstract

A method of parsing a natural language phrase to retrieve data associated with a power monitoring system. An input receives a natural language phrase including terms, one of which indicates a data manipulation action for manipulating data of the power monitoring system. A parser component parses the phrase to extract the terms and maps those terms to corresponding category items. The parser component constructs a structured query based on the category items and provides the query to a processor component that retrieves the data from a data source specified in the submitted phrase. The processor component also manipulates the data according to the data manipulation action. The output from the processor component is provided to a presentation component that formats the manipulated data for presentation on a video display, and the formatted manipulated data is displayed on the video display.

Claims

1. A method of parsing a natural language phrase to retrieve data associated with a power monitoring system, comprising:

receiving a natural language phrase including one or more terms composed of alphanumeric characters, the terms including at least one term that indicates a data manipulation action for manipulating data of the power monitoring system, the data having a data type;

parsing the phrase to extract the one or more terms;

mapping each of the one or more terms to one or more category items, wherein the category items include a data quantity category indicating a quantity of data representing one or more electrical characteristics measured by a monitoring point in the power monitoring system, a data source category indicating a source of the data measured by the monitoring point, and a selection criteria category indicating the data manipulation action to be performed on the data measured by the monitoring point;

constructing a structured query based on the one or more category items;

responsive to receiving the structured query, retrieving the data from the data source mapped to the data source category and manipulating the data according to the data manipulation action mapped to the selection criteria category to produce manipulated data;

formatting at least the manipulated data for presentation on a video display to produce formatted manipulated data; and

displaying the formatted manipulated data on the video display.

2. The method of claim 1, wherein the manipulating the data according to the data manipulation action includes generating a statistical summary of the data.

3. The method of claim 2, wherein the generating the statistical summary of the data includes locating a minimum, maximum, or average value in the data.

4. The method of claim 1, wherein the manipulating the data according to the data manipulation action includes organizing the data into a set of data over a time range.

5. The method of claim 1, wherein the manipulating the data according to the data manipulation action includes performing a calculation on the data.

6. The method of claim 1, further comprising automatically suggesting, as the phrase is being entered, at least one suggested term based on at least one criterion.

7. The method of claim 6, wherein the at least one criterion is a frequency of use of past terms that have already been entered.

8. The method of claim 6, wherein the at least one criterion is a chronological order of past terms that have already been entered.

9. The method of claim 6, wherein the at least one criterion is an alphabetical order of past terms that have already been entered.

10. The method of claim 6, wherein the at least one criterion is a correlation between the term being entered and the past terms that have already been entered.

11. The method of claim 10, wherein the at least one suggested term is a plurality of suggested terms, the method further comprising indicating a strength of the correlation by displaying each of the plurality of suggested terms in respective colors indicative of the respective correlation strengths.

12. The method of claim 1, wherein the one or more functional modules include a real-time data module that processes data representing the one or more electrical characteristics in real-time as the one or more electrical characteristic is being measured by the monitoring point in the power monitoring system.

13. The method of claim 1, wherein the one or more functional modules include a historical trend data module that processes data representing the one or more electrical characteristics over a predetermined historical time period as the one or more electrical characteristics were measured by the monitoring point in the power monitoring system.

14. The method of claim 1, wherein the one or more functional modules include an alarm module that processes one or more alarm events indicating whether the at least one or more electrical characteristics measured by the monitoring point in the power monitoring system exceeds a predetermined alarm threshold.

15. The method of claim 1, wherein the one or more functional modules include a waveform data module that processes waveform capture data representing one or more electrical characteristics measured by the monitoring point in the power monitoring system.

16. The method of claim 1, wherein the one or more functional modules includes a diagnostics module that processes information indicative of a diagnostic criterion for analyzing data measured by one or more power monitoring devices in the power monitoring system.

17. The method of claim 1, wherein the one or more functional modules includes a power quality event module that retrieves a range of power quality event data from a requested data source in the power monitoring system.

18. The method of claim 1, wherein the at least one or more electrical characteristics include voltage, current, harmonic magnitudes, power factor, vars, volt-amps, frequency, power demand, or energy demand.

19. The method of claim 1, wherein the monitoring point is a physical monitoring device or a virtual monitoring point in the power monitoring system, wherein the virtual monitoring point derives monitored data from other monitoring points in the power monitoring system.

20. A system for parsing a natural language phrase to retrieve data associated with a power monitoring system, comprising:

an input that receives natural language phrase including one or more terms composed of alphanumeric characters, the terms including at least one term that indicates a data manipulation action for manipulating data of the power monitoring system, the data having a data type;

a parser component that;

parses the phrase to extract the one or more terms,

maps each of the one or more terms to one or more category items, wherein the category items include a data quantity category indicating a quantity of data representing one or more electrical characteristics measured by a monitoring point in the power monitoring system, a data source category indicating a source of the data measured by the monitoring point, and a selection criteria category indicating the data manipulation action to be performed on the data measured by the monitoring point, and constructs a structured query based on the one or more category items;

a processor component that, responsive to receiving the structured query, retrieves the data from the data source mapped to the data source category and manipulates the data according to the data manipulation action mapped to the selection criteria category to produce manipulated data;

a presentation component that formats the manipulated data for presentation on a video display to produce formatted manipulated data; and

an output that displays the formatted manipulated data on the video display.