METHOD AND APPARATUS FOR ANALYZING WAVEFORM SIGNALS OF A POWER SYSTEM

• US 20110213577A1
• Filed: 03/25/2009
• Published: 09/01/2011
• Est. Priority Date: 03/25/2008
• Status: Active Grant

• Current Assignee: ABB Power Grids Switzerland AG (ABB Limited), ABB Schweiz AG (ABB Limited)
• Original Assignee: ABB Research Ltd. (ABB Ltd.)
• Inventors: Mousavi, Mirrasoul, Maharsi, Mohamed Y. Haj
• Granted Patent: US 10,281,504 B2
• Time in Patent Office: Days
• US Class Current: 702/66
• CPC Class Codes: G01R 19/2509 Details concerning sampling, digitizing or waveform capturing
• CPC Class Codes: G01R 19/2513 Arrangements for monitoring electric power systems, e.g. power lines or loads; Logging
• CPC Class Codes: G05B 23/0221 Preprocessing measurements, e.g. data collection rate adjustment; Standardization of measurements; Time series or signal analysis, e.g. frequency analysis or wavelets; Trustworthiness of measurements; Indexes therefor; Measurements using easily measured parameters to estimate parameters difficult to measure; Virtual sensor creation; De-noising; Sensor fusion; Unconventional preprocessing inherently present in specific fault detection methods like PCA-based methods
• CPC Class Codes: G06F 17/18 for evaluating statistical data , e.g. average values, frequency distributions, probability functions, regression analysis forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04

Abstract

Method and apparatus for analyzing waveform signals of a power system. A waveform signal obtained from the power system is digitalized and then filtered. One or more segments are identified on the filtered waveform signal and, based on the information content of the identified segments, each of the one or more identified segments are classified as corresponding to a normal or abnormal condition occurred in the power system.

Claims

1. A method for analyzing waveform signals of a power system, comprising:

a) digitalizing a waveform signal obtained from said power system;

b) filtering the digitalized waveform signal;

c) identifying one or more segments on said filtered waveform signal; and

d) based on the information content of each of said one or more identified segments, classifying each of said one or more identified segments as corresponding to a normal or abnormal condition occurred in the power system.

2. The method of claim 1 wherein said step b) comprises re-sampling said digitalized waveform signal.

3. The method of claim 1 wherein, said step b) comprises removing DC components and noise from said digitalized waveform signal.

4. The method of claim 1 wherein said step b) comprises calculating the Root Mean Square (RMS) signal for the digitalized waveform signal over a moving window of fixed size.

5. The method of claim 1 wherein said step c) comprises determining change points on said filtered waveform signal.

6. The method of claim 1 wherein said change points are determined relative to a change parameter without target value.

7. The method of claim 5 wherein a change point is detected when the value calculated for said change parameter is equal to or higher than a predefined alarm value.

8. The method of claim 5 wherein, said step c) comprises grouping the determined change points forming clusters whose respective boundaries delimit approximate segments of the filtered waveform signal.

9. The method of claim 8 wherein clusters are formed each time the difference between successive identified change points is above a predefined cut-off value.

10. The method of claim 9 wherein it further comprises validating the formed clusters by using different values for one or more of said cut-off value, alarm value, and threshold value.

11. The method of claim 9 wherein said step c) comprises fine-tuning the boundaries of the formed clusters by applying an unsupervised classification technique to said digitalized waveform signal.

12. The method of claim 11 wherein said-fine tuning is performed based on the pattern of the digitalized waveform signal.

13. An intelligent electronic device suitable to be connected to a power system, said intelligent electronic device comprising a computer device having code therein configured to:

a) digitalize a waveform signal obtained from said power system;

b) filter the digitalized waveform signal;

c) identify one or more segments on said filtered waveform signal; and

d) based on the information content of each of said one or more identified segments, classify each of said one or more identified segments as corresponding to a normal or abnormal condition occurred in the power system.

14. The intelligent electronic device of claim 13, wherein said computer device has code therein configured to:

re-sample said digitalized waveform signal;

remove DC components and noise from said digitalized waveform signal.

15. The intelligent electronic device of claim 13, wherein said computer device has code therein configured to calculate the RMS signal for the digitalized waveform signal over a moving window of fixed size.

16. The intelligent electronic device of claim 13, wherein said computer device has code therein configured to determine change points on said filtered waveform signal, wherein said change points are determined relative to a change parameter without target value and a change point is detected when the value calculated for said change parameter is equal to or higher than a predefined alarm value.

17. The intelligent electronic device of claim 16, wherein said computer device has code therein configured to group the determined change points forming clusters whose respective boundaries delimit approximate segments of the filtered waveform signal, clusters being formed each time the difference between successive identified change points is above a predefined cut-off value.

18. The intelligent electronic device of claim 17, wherein said computer device has code therein configured to validate the formed clusters.

19. The intelligent electronic device of claim 17, wherein said computer device has code therein configured to fine-tune the boundaries of the formed clusters, said-fine tuning being performed based on the pattern of the digitalized waveform signal.

20. A computer readable medium for analyzing waveform signals of a power system, said computer-readable medium having computer-readable instructions stored thereon for execution by a microprocessor to perform a method comprising:

a) digitalizing a waveform signal obtained from said power system;

b) filtering the digitalized waveform signal;

c) identifying one or more segments on said filtered waveform signal; and

d) based on the information content of each of said one or more identified segments, classifying each of said one or more identified segments as corresponding to a normal or abnormal condition occurred in the power system.