초록 ▼

A method and an apparatus detect series and/or parallel arc faults in AC and DC systems. The method according to one embodiment inputs an AC current signal; extracts a fundamental component of the AC current signal and monitoring an amplitude variation profile for the fundamental component, thereby

A method and an apparatus detect series and/or parallel arc faults in AC and DC systems. The method according to one embodiment inputs an AC current signal; extracts a fundamental component of the AC current signal and monitoring an amplitude variation profile for the fundamental component, thereby generating a first arc fault detection measure; detects non-stationary changes in the AC current signal applying at least one measure of order higher than one, thereby generating a second arc fault detection measure; and determines whether an arc fault exists based on the first arc fault detection measure and the second arc fault detection measure.

대표청구항 ▼

We claim: 1. A method of detecting series and/or parallel arc faults in an AC or DC system, said method comprising: inputting a current signal; extracting a component of said current signal and monitoring an amplitude variation profile for said component, thereby generating a first arc fault detect

We claim: 1. A method of detecting series and/or parallel arc faults in an AC or DC system, said method comprising: inputting a current signal; extracting a component of said current signal and monitoring an amplitude variation profile for said component, thereby generating a first arc fault detection measure; detecting non-stationary changes in said current signal by applying at least one statistical or probabilistic based measure of order higher than one, thereby generating a second arc fault detection measure; and determining whether an arc fault exists based on said first arc fault detection measure and said second arc fault detection measure. 2. The method of detecting arc faults in an AC or DC system as recited in claim 1, wherein said current signal is an AC current signal, said component is a fundamental component, and said extracting step includes: adapting to frequency changes of said AC current signal using a phase locked loop; applying a moving/sliding discrete Fourier transform (DFT) to extract said fundamental component of said AC current signal in real time. 3. The method of detecting arc faults in an AC or DC system as recited in claim 1, wherein said current signal is an AC current signal, said component is a fundamental component, and said amplitude variation profile for said component is determined by counting a number of fundamental component maxima within a selected time interval and comparing said number to a threshold. 4. The method of detecting arc faults in an AC or DC system as recited in claim 1, wherein said current signal is an AC current signal, said component is a fundamental component, and said step of detecting non-stationary changes in said AC current signal includes: comb filtering said AC current signal to suppress fundamental and harmonic components of said signal; and normalizing the result of said comb filtering step with respect to the fundamental amplitude. 5. The method of detecting arc faults in an AC or DC system as recited in claim 1, wherein said step of detecting non-stationary changes in said current signal detects transients, inter-harmonics, and high frequency component variations, and applies high order statistical analysis to measure skewness and kurtosis and detect non-stationary and wideband changes in said current signal. 6. The method of detecting arc faults in an AC or DC system as recited in claim 1, wherein said component is a DC component of said current signal. 7. The method of detecting arc faults in an AC or DC system as recited in claim 6, further comprising: applying a moving/sliding discrete Fourier transform (DFT) to extract said DC component of said current signal in real time. 8. The method of detecting arc faults in an AC or DC system as recited in claim 6, wherein said amplitude variation profile for said DC component is determined by counting a number of maxima within a selected time interval and comparing said number to a threshold. 9. The method of detecting arc faults in an AC or DC system as recited in claim 6, wherein said step of detecting non-stationary changes in said current signal includes: comb filtering said current signal; and normalizing the result of said comb filtering step. 10. The method of detecting arc faults in an AC or DC system as recited in claim 6, wherein said step of detecting non-stationary changes in said current signal detects transients, inter-harmonics, and high fteciuency component variations, and applies high order statistical analysis to measure skewness and kurtosis and detect non-stationary and wideband changes in said current signal. 11. The method of detecting arc faults in an AC or DC system as recited in claim 1, wherein said statistical or probabilistic based measure of order higher than one is a High Order Spectral measure. 12. The method of detecting arc faults in an AC or DC system as recited in claim 1, wherein said step of detecting non-stationary changes in said current signal applies high order statistical analysis to measure skewness and kurtosis and detect non-stationary and wideband changes in said current signal. 13. An apparatus for detecting series and/or parallel arc faults in an AC or DC system, said apparatus comprising: an input unit for inputting a current signal; a steady state component analysis unit for extracting a component of said current signal and monitoring an amplitude variation profile for said component, thereby generating a first arc fault detection measure; a non-steady state component analysis unit for detecting non-stationary changes in said current signal by applying at least one statistical or probabilistic based measure of order higher than one, thereby generating a second arc fault detection measure; and a detection unit for determining whether an arc fault exists based on said first arc fault detection measure and said second arc fault detection measure. 14. The apparatus for detecting arc faults in an AC or DC system according to claim 13, wherein said current signal is an AC current signal, said component is a fundamental component of said AC current signal, said steady state component analysis unit extracts said fundamental component of said AC current signal by adapting to frequency changes of said AC current signal using a phase locked loop; and applying a moving/sliding discrete Fourier transform (DFT) to extract said fundamental component of said AC current signal in real time. 15. The apparatus for detecting arc faults in an AC or DC system according to claim 13, wherein said current signal is an AC current signal, said component is a fundamental component, and said steady state component analysis unit determines said amplitude variation profile for said fundamental component by counting a number of fundamental component maxima within a selected time interval and comparing said number to a threshold. 16. The apparatus for detecting arc faults in an AC or DC system according to claim 13, wherein said current signal is an AC current signal, said component is a fundamental component, and said non-steady state component analysis unit detects non-stationary changes in said AC current signal by comb filtering said AC current signal to suppress fundamental and harmonic components of said signal; and normalizing the result of said comb filtering step with respect to the fundamental amplitude. 17. The apparatus for detecting arc faults in an AC or DC system according to claim 13, wherein said current signal is an AC current signal, said component is a fundamental component, and said non-steady state component analysis unit detects transients, inter-harmonics, and high frequency component variations in said AC current signal, and said non-steady state component analysis unit detects non-stationary changes in said AC current signal by applying high order statistical analysis to measure skewness and kurtosis and detect non-stationary and wideband changes in said AC current signal. 18. The apparatus for detecting arc faults in an AC or DC system according to claim 13, wherein said component is a DC component of said current signal. 19. The apparatus for detecting arc faults in an AC or DC system according to claim 18, wherein said steady state component analysis unit extracts a DC component of said current signal by applying a moving/sliding discrete Fourier transform (DFT) to extract said DC component of said current signal in real time. 20. The apparatus for detecting arc faults in an AC or DC system according to claim 18, wherein said steady state component analysis unit determines said amplitude variation profile for said DC component by counting a number of maxima within a selected time interval and comparing said number to a threshold. 21. The apparatus for detecting arc faults in an AC or DC system according to claim 18, wherein said non-steady state component analysis unit detects non-stationary changes in said current signal by comb filtering said current signal; and normalizing the result of said comb filtering step. 22. The apparatus for detecting arc faults in an AC or DC system according to claim 18, wherein said non-steady state component analysis unit detects transients, inter-harmonics, and high frequency component variations in said current signal, and detects non-stationary changes in said current signal by applying high order statistical analysis to measure skewness and kurtosis and detect non-stationary and wideband changes in said current signal. 23. The apparatus for detecting arc faults in an AC or DC system according to claim 18, wherein said apparatus is applied in an aerospace power distribution system. 24. The apparatus for detecting arc faults in an AC or DC system according to claim 13, wherein said apparatus is applied in an aerospace power distribution system. 25. The apparatus for detecting arc faults in an AC or DC system according to claim 13, wherein said statistical or probabilistic based measure of order higher than one is a High Order Spectral measure. 26. The apparatus for detecting arc faults in an AC or DC system according to claim 13, wherein said non-steady state component analysis unit detects non-stationary changes in said current signal by applying high order statistical analysis to measure skewness and kurtosis and detect non-stationary and wideband changes in said current signal.