Systems, methods, and device for arc fault detection

• US 20050286185A1
• Filed: 06/16/2005
• Published: 12/29/2005
• Est. Priority Date: 06/21/2004
• Status: Active Grant

• Current Assignee: Siemens Industry Incorporated (Siemens AG)
• Original Assignee: Siemens Energy & Automation Incorporated (Siemens AG)
• Inventors: Restrepo, Carlos, Henson, Jeremy
• Granted Patent: US 7,307,820 B2
• Time in Patent Office: Days
• US Class Current: 361/42
• CPC Class Codes: H02H 1/0015 Using arc detectors

Abstract

Certain exemplary embodiments described herein comprise a method comprising a plurality of activities comprising: discriminating an arcing condition from a non-arcing broadband signal, and, in response to detecting an arcing condition, outputting a trip signal to a circuit breaker controlling a predetermined alternating current circuit associated with the arcing condition.

Claims

1. A method comprising:

sensing a broadband RF signal within an alternating current signal transmitted on a predetermined alternating current circuit;

down-converting the sensed broadband RF signal to an intermediate frequency;

generating a received signal strength indication signal representative of a signal strength of the sensed broadband RF signal at the intermediate frequency;

sampling the received signal strength indication signal at a predetermined sample frequency;

repeatedly, for a predetermined one of a predetermined number of sequential time intervals, each time interval corresponding to a predetermined number of sequential received signal strength indication signal samples, each sample from the predetermined number of samples assigned a sample identifier, for each sample identifier, determining a maximum amplitude of all samples corresponding to that sample identifier in all of the predetermined number of sequential time intervals and storing that maximum amplitude;

repeatedly, for the predetermined number of time intervals, for each sample identifier, determining a mean amplitude of all stored maximum amplitudes associated with the predetermined number of time intervals and storing that mean amplitude;

repeatedly, for a computed signal comprised of each of the stored mean amplitudes, detecting a predetermined number of disturbances, each disturbances characterized by a steep rising edge and a steep falling edge;

characterizing the broadband RF signal as related to an arcing condition if a gap between a steep falling edge associated with a disturbance of a first predetermined number and a steep rising edge associated with a disturbance of a second predetermined number exceeds a predetermined value;

opening the predetermined alternating current circuit if the characterized broadband RF signal is related to an arcing condition.

2. A method comprising:

repeatedly, for a predetermined one of a predetermined number of sequential time intervals, each time interval corresponding to a predetermined number of sequential samples of an indication signal related to a sensed broadband signal within an alternating current signal transmitted on a predetermined alternating current circuit, each sample from the predetermined number of samples assigned a sample identifier, for each sample identifier, determining a statistical amplitude measure of all samples corresponding to that sample identifier in all of the predetermined number of sequential time intervals and storing that statistical amplitude measure;

repeatedly, for the predetermined number of time intervals, for each sample identifier, determining a mean amplitude value of all stored statistical amplitude measures associated with the predetermined number of time intervals and storing that mean amplitude value;

for a computed signal comprised of each of the stored mean amplitude values, detecting an arcing condition; and

outputting a trip signal to a circuit breaker controlling the predetermined alternating current circuit.

3. The method of claim 2, further comprising:

sensing the broadband signal.

4. The method of claim 2, further comprising:

down-converting the sensed broadband signal to a predetermined intermediate frequency.

5. The method of claim 2, further comprising:

generating a received signal strength indication signal representative of a signal strength of the sensed broadband signal at a predetermined frequency.

6. The method of claim 2, further comprising:

sampling the indication signal.

7. The method of claim 2, further comprising:

sampling the indication signal at a predetermined sampling frequency.

8. The method of claim 2, further comprising:

discriminating a BPL signal.

9. The method of claim 2, further comprising:

detecting a predetermined disturbance in the computed signal.

10. The method of claim 2, further comprising:

detecting a duration of a predetermined disturbance in the computed signal.

11. The method of claim 2, further comprising:

detecting a predetermined number of disturbances in the computed signal.

12. The method of claim 2, further comprising:

detecting a rising edge of a disturbance in the computed signal.

13. The method of claim 2, further comprising:

detecting a falling edge of a disturbance in the computed signal.

14. The method of claim 2, further comprising:

detecting a slope of the computed signal.

15. The method of claim 2, further comprising:

detecting a duration between predetermined disturbances in the computed signal.

16. The method of claim 2, further comprising:

detecting a duration of a gap between predetermined disturbances in the computed signal.

17. The method of claim 2, further comprising:

comparing a gap between predetermined disturbances in the computed signal to a predetermined criteria.

18. The method of claim 2, further comprising:

scoring a gap between predetermined disturbances in the computed signal.

19. The method of claim 2, further comprising:

causing the predetermined alternating current circuit to trip.

20. The method of claim 2, wherein for each sample identifier, the statistical amplitude measure is an average amplitude value of all samples corresponding to that sample identifier for the predetermined number of time intervals.

21. The method of claim 2, wherein for each sample identifier, the statistical amplitude measure is a maximum amplitude value of all samples corresponding to that sample identifier for the predetermined number of time intervals.

22. The method of claim 2, wherein the broadband signal is a radio frequency signal.

23. A system comprising:

a statistical amplitude processor module adapted to, repeatedly, for a predetermined one of a predetermined number of sequential time intervals, each time interval corresponding to a predetermined number of sequential samples of a indication signal related to a sensed broadband signal carried on an alternating current signal transmitted on a predetermined alternating current circuit, each sample from the predetermined number of sequential samples assigned a sample identifier, for each sample identifier, determine a statistical amplitude measure of all samples corresponding to that sample identifier in all of the predetermined number of sequential time intervals and storing that statistical amplitude measure;

a mean amplitude processor module adapted to, repeatedly, for the predetermined number of time intervals, for each sample identifier, determine a mean amplitude value of all stored statistical amplitude measures associated with the predetermined number of time intervals and storing that mean amplitude value;

a detector adapted to, for a computed signal comprised of each of the stored mean amplitude values, detect an arc event; and

a signal generator adapted to output a trip signal to a circuit breaker controlling the predetermined alternating current circuit.