IPC분류정보
국가/구분 |
United States(US) Patent
등록
|
국제특허분류(IPC7판) |
|
출원번호 |
US-0166937
(2002-06-11)
|
발명자
/ 주소 |
- Hu, John Zhongzhi
- Vora, Nishith Pramod
- Goodman, George Charles
|
출원인 / 주소 |
|
대리인 / 주소 |
|
인용정보 |
피인용 횟수 :
2 인용 특허 :
2 |
초록
▼
A method of detecting faults in a fuel cell system, the method comprising: comparing a plurality of fuel cell measurements to respective ones of a plurality of range limits to yield a plurality of range flags; differentiating the fuel cell measurements to yield a plurality of fuel cell rates; compar
A method of detecting faults in a fuel cell system, the method comprising: comparing a plurality of fuel cell measurements to respective ones of a plurality of range limits to yield a plurality of range flags; differentiating the fuel cell measurements to yield a plurality of fuel cell rates; comparing the fuel cell rates to respective ones of a plurality of rate limits to yield a plurality of rate flags; and classifying the range flags and the rate flags to yield a fault detection decision.
대표청구항
▼
1. A method of detecting faults in a fuel cell system, said method comprising:comparing a plurality of fuel cell measurements to respective ones of a plurality of range limits to yield a plurality of range flags;differentiating said fuel cell measurements to yield a plurality of fuel cell rates;comp
1. A method of detecting faults in a fuel cell system, said method comprising:comparing a plurality of fuel cell measurements to respective ones of a plurality of range limits to yield a plurality of range flags;differentiating said fuel cell measurements to yield a plurality of fuel cell rates;comparing said fuel cell rates to respective ones of a plurality of rate limits to yield a plurality of rate flags; andclassifying said range flags and said rate flags to yield a fault detection decision. 2. The method of claim 1 wherein said fuel cell measurements comprise at least one quantity selected from a group consisting of voltage, temperature, power, current, pressure, gas flow, and concentration. 3. The method of claim 1 further comprising dynamic filtering a plurality of unfiltered fuel cell measurements to yield said fuel cell measurements. 4. The method of the previous claim wherein said step of dynamic filtering comprises at least one method selected from a group consisting of lowpass filtering, bandpass filtering, highpass filtering, Kalman filtering, and rank filtering. 5. The method of claim 1 further comprising:computing a residual statistic from said fuel cell measurements; andcomparing said residual statistic to a residual threshold to yield a residual flag,said step of classifying comprising using said residual flag. 6. The method of the previous claim wherein said step of computing comprises a method selected from a group consisting of physical redundancy, analytical redundancy, and Kalman filtering. 7. A method of detecting faults in a fuel cell system, said method comprising:dynamic filtering a plurality of unfiltered fuel cell measurements to yield a plurality of fuel cell measurements;comparing said fuel cell measurements to respective ones of a plurality of range limits to yield a plurality of range flags;differentiating said fuel cell measurements to yield a plurality of fuel cell rates;comparing said fuel cell rates to respective ones of a plurality of rate limits to yield a plurality of rate flags;computing a residual statistic from said fuel cell measurements;comparing said residual statistic to a residual threshold to yield a residual flag; andclassifying said range flags, said rate flags, and said residual flag to yield a fault detection decision. 8. The method of claim 7 wherein said unfiltered fuel cell measurements comprise at least one quantity selected from a group consisting of voltage, temperature, power, current, pressure, gas flow, and concentration. 9. The method of claim 7 wherein said step of dynamic filtering comprises at least one method selected from a group consisting of lowpass filtering, bandpass filtering, highpass filtering, Kalman filtering, and rank filtering. 10. The method of claim 7 wherein said step of computing comprises a method selected from a group consisting of physical redundancy, analytical redundancy, and Kalman filtering. 11. An apparatus comprising:a fuel cell;a range limit comparator adapted for comparing a plurality of fuel cell measurements, from said fuel cell, to respective ones of a plurality of range limits to yield a plurality of range flags;a differentiator adapted for differentiating said fuel cell measurements to yield a plurality of fuel cell rates;a rate limit comparator adapted for comparing said fuel cell rates to respective ones of a plurality of rate limits to yield a plurality of rate flags; anda fault classifier adapted for classifying said range flags and said rate flags to yield a fault detection decision. 12. The apparatus of claim 11 wherein said fuel cell measurements comprise at least one quantity selected from a group consisting of voltage, temperature, power, current, pressure, gas flow, and concentration. 13. The apparatus of claim 11 further comprising a dynamic filter adapted for dynamic filtering a plurality of unfiltered fuel cell measurements from said fuel cell to yield said fuel cell measurements. 14. The apparatus of the previous cl aim wherein said dynamic filter implements at least one method selected from a group consisting of lowpass filtering, bandpass filtering, highpass filtering, Kalman filtering, and rank filtering. 15. The apparatus of claim 11 further comprising:a residual computer adapted for computing a residual statistic from said fuel cell measurements; anda residual threshold comparator adapted for comparing said residual statistic to a residual threshold to yield a residual flag,said fault classifier using said residual flag. 16. The apparatus of the previous claim wherein said residual computer is further adapted to implement at least one method selected from a group consisting of physical redundancy, analytical redundancy, and Kalman filtering. 17. An apparatus comprising:a fuel cell;a dynamic filter adapted for dynamic filtering a plurality of unfiltered fuel cell measurements from said fuel cell to yield a plurality of filtered fuel cell measurements;a range limit comparator adapted for comparing said filtered fuel cell measurements to respective ones of a plurality of range limits to yield a plurality of range flags;a differentiator adapted for differentiating said filtered fuel cell measurements to yield a plurality of fuel cell rates;a rate limit comparator adapted for comparing said fuel cell rates to respective ones of a plurality of rate limits to yield a plurality of rate flags;a residual computer adapted for computing a residual statistic from said filtered fuel cell measurements;a residual threshold comparator adapted for comparing said residual statistic to a residual threshold to yield a residual flag; anda fault classifier adapted for classifying said range flags, said rate flags, and said residual flag to yield a fault detection decision. 18. The apparatus of claim 17 wherein said unfiltered fuel cell measurements comprise at least one quantity selected from a group consisting of voltage, temperature, power, current, pressure, gas flow, and concentration. 19. The apparatus of the claim 17 wherein said dynamic filter implements at least one method selected from a group consisting of lowpass filtering, bandpass filtering, highpass filtering, Kalman filtering, and rank filtering. 20. The apparatus of the claim 17 wherein said residual computer is further adapted to implement at least one method selected from a group consisting of physical redundancy, analytical redundancy, and Kalman filtering.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.