Statistical method and apparatus for monitoring parameters in an electric power distribution system

• US 20050273280A1
• Filed: 06/03/2004
• Published: 12/08/2005
• Est. Priority Date: 06/03/2004
• Status: Active Grant

• Current Assignee: Eaton Intelligent Power Limited (Eaton Corporation PLC.)
• Original Assignee: Eaton Corp. (Wisconsin) (Eaton Corporation PLC.)
• Inventors: Cox, Roger W.
• Granted Patent: US 7,054,769 B2
• Time in Patent Office: Days
• US Class Current: 702/60
• CPC Class Codes: G01R 19/2513 Arrangements for monitoring electric power systems, e.g. power lines or loads; Logging
• CPC Class Codes: G01R 21/133 by using digital technique

Abstract

Dynamic thresholds statistically derived are used to define normal conditions at a specific site in an electric power distribution system. High and low and/or caution and alarm thresholds for various parameters are generated as multiples of the standard deviation of the long-term average of measured values calculated from samples gathered over repetitive intervals, such as ten minutes. Weekly averages are used until sufficient data has been gathered to maintain a running one year average.

Claims

1. A method for monitoring parameters in an electric power distribution system comprising the steps of:

measuring a selected parameter of the electric power distribution system repetitively to generate measured values of the selected parameter;

periodically determining the standard deviation of the measured values;

periodically establishing at least one dynamic threshold for the measured values as a function of the standard deviation; and

generating an output relating the measured values to the at least one dynamic threshold.

2. The method of claim 1 wherein measuring the selected parameter repetitively comprises taking multiple samples of the selected parameter over a selected interval and generating the measured value as a function of the multiple samples.

3. The method of claim 2 wherein the measured values are generated as an average function of the multiple samples over the selected time interval.

4. The method of claim 1 wherein periodically generating the standard deviation comprises periodically generating a long-term average of the measured values and using the long-term average to generate the standard deviation.

5. The method of claim 4 wherein generating a long-term average comprises generating a first moving average of the measured values over a first time period, generating a second moving average over a second time period which is a multiple of the first time period using multiples of the first moving average, and generating the long-term average using only the first moving average until the method has been employed for the second time period and thereafter generating the long-term average using the second moving average.

6. The method of claim 4 wherein the at least one dynamic threshold is generated by adding a multiplier times the standard deviation to the long-term average.

7. The method of claim 6 wherein the multiplier ranges from about three to about six.

8. The method of claim 6 wherein the at least one dynamic threshold comprises a low dynamic threshold generated with a negative multiplier, and a high dynamic threshold generated with a positive multiplier.

9. The method of claim 1 wherein generating the at least one dynamic threshold is generated within a range of about three to about six times the standard deviation.

10. The method of claim 9 comprising generating a first dynamic threshold at about three times the standard deviation and a second dynamic threshold at about six times the standard deviation.

11. The method of claim 10 wherein generating the output comprises generating a caution when the measured value reaches the first dynamic threshold and generating at least one of an alarm and an action when the measured value reaches the second dynamic threshold.

12. The method of claim 1 wherein generating an output comprises generating a response when the measured value reaches the dynamic threshold.

13. The method of claim 1 wherein the selected parameter is selected from a list comprising:

volts, amps, vars, watts, VA, power factor, total harmonic distortion in the current, total harmonic distortion in the voltage, flicker, and frequency.

14. A monitoring system for monitoring an electric power distribution system, comprising:

a sensor sensing conditions in the electric power distribution system;

processing means generating from sensed conditions measured values of a selected parameter and periodically generating a standard deviation of the measured values and at least one dynamic threshold for the measured values using the standard deviation; and

output means generating an output relating the measured values to the at least one dynamic threshold.

15. The monitoring system of claim 14 wherein the processing means comprises means generating a long-term average of the measured values and generating a low dynamic threshold by subtracting a function of the standard deviation from the long-term average and generating a high dynamic threshold by adding the function of the standard deviation to the long-term average.

16. The monitoring system of claim 14 wherein the output means comprises means generating a response when the measured value reaches the at least one dynamic threshold.

17. The monitoring system of claim 14 wherein the processing means comprises means generating samples of the selected parameter over a selected interval and generating the measured values as an average of the samples over the selected interval.

18. The monitoring system of claim 14 wherein the processing means comprises means generating a long-term average of the measured values, and means generating the at least one dynamic threshold by adding a multiplier times the standard deviation to the long-term average.

19. The monitoring system of claim 18 wherein the multiplier is positive to generate a high dynamic threshold and negative to generate a low dynamic threshold.

20. The monitoring system of claim 18 wherein the means generating the at least one dynamic threshold generates a first dynamic threshold using a first multiplier and a second dynamic threshold using a second multiplier larger than the first multiplier, and wherein the output means generates a caution indication when the first dynamic threshold is reached by the measured values and generates at least one of a warning and an action when the second dynamic threshold is reached by the measured values.