Method and apparatus for estimating an actual magnitude of a physical parameter on the basis of three or more redundant signals

• US 6,073,262 A
• Filed: 05/30/1997
• Issued: 06/06/2000
• Est. Priority Date: 05/30/1997
• Status: Expired due to Term

• Current Assignee: United Technologies Corporation (Raytheon Technologies Corporation d/b/a RTX)
• Original Assignee: United Technologies Corporation (Raytheon Technologies Corporation d/b/a RTX)
• Inventors: Larkin, Louis J., Healy, Timothy A., Kerr, Laura J.
• Primary Examiner(s): An, Meng-Ai T.
• Assistant Examiner(s): EL HADY, NABIL M
• Application Number:
  US08/867,702
• Time in Patent Office: 1,103 Days
• Field of Search: 701/27, 701/144, 701/57, 701/106, 701/4, 701/11, 701/117, 702/181, 702/199, 706/45, 376/245, 60/39.281, 73/116, 714/2, 714/6, 714/25, 714/26, 714/31, 714/33, 714/48, 714/723, 714/736, 714/737, 714/786, 714/794, 714/797, 714/819
• US Class Current: 714/736
• CPC Class Codes: G01D 3/08 with provision for safeguarding the apparatus, e.g. against abnormal operation, against breakdown
• CPC Class Codes: G05B 13/0275 using fuzzy logic only
• CPC Class Codes: G05B 23/0254 based on a quantitative model, e.g. mathematical relationships between inputs and outputs; functions: observer, Kalman filter, residual calculation, Neural Networks
• CPC Class Codes: G05B 9/03 with multiple-channel loop, i.e. redundant control systems

Abstract

An apparatus includes at least three redundant signals each indicative of a sensed magnitude of a parameter. Parity space and fuzzy logic techniques may be used to provide a signal representing an estimate of the actual magnitude of the parameter. A method for generating an estimate of an actual magnitude of a parameter provides at least three redundant signals may use parity space and fuzzy logic to produce a signal representing an estimate of the actual magnitude of the parameter.

Claims

1. An apparatus for use in estimating an actual magnitude of a physical parameter, where at least three redundant signals are input to the apparatus, each input signal is indicative of a sensed magnitude of the parameter, the apparatus comprising:

a differencer providing in response to the at least three redundant input signals at least three difference signals indicative of the relative differences between the sensed magnitudes;

an evaluator for computing in response to said at least three difference signals, a plurality of evaluation signals; and

a synthesizer for generating in response to said evaluation signals, an estimate signal indicative of an estimate of the actual magnitude of the physical parameter,

wherein said estimate signal has a magnitude indicative of a summation of the product of the magnitude indicated by each redundant input signal and an associated proportionality factor, wherein each proportionality factor has a value equal to one of a plurality of possible values for that proportionality factor, the value of each proportionality factor depending upon said magnitudes of said difference signals.

2. The apparatus of claim 1 wherein said possible values of said proportionality factors ranges from substantially zero to substantially unity.

3. The apparatus of claim 1 wherein said possible values of said proportionality factors comprises a generally continuous set of values.

4. The apparatus of claim 1 wherein each of said difference signals has a magnitude proportional to the difference between the sensed magnitudes indicated by two of the redundant input signals.

5. The apparatus of claim 1 wherein said estimate signal has a magnitude that is described by a smoothly varying function.

6. The apparatus as in claim 1 wherein said evaluation signals are of a type indicative of weighting factors and said estimate signal has a magnitude indicative of a weighted average based on said weighting factors.

7. The apparatus as in claim 6 wherein said evaluator further receives the redundant input signals and provides in response to the redundant input signals a plurality of signals each indicative of a possible estimate of the actual magnitude of the physical parameter, and wherein said synthesizer receives said plurality of signals and provides said estimate signal which is indicative of a weighted average of said plurality of signals.

8. An apparatus for use in estimating an actual magnitude of a physical parameter, where at least three redundant signals are input to the apparatus, each input signal is indicative of a sensed magnitude of the parameter, the apparatus comprising:

a differencer providing in response to the at least three redundant input signals at least three difference signals indicative of the relative differences between the sensed magnitudes;

an evaluator for computing a plurality of evaluation signals in response to an input of said at least three difference signals and the at least three redundant input signals,

wherein said evaluation provides said evaluation signals that are representative of fuzzy logic signals; and

a synthesizer for generating in response to said evaluation signals, an estimate signal indicative of an estimate of the actual magnitude of the physical parameter.

9. The apparatus of claim 8 wherein said evaluator comprises a fuzzy inferencer having a rulebase with a plurality of rules, each of said rules having a condition portion and a consequent portion, and wherein said consequent portion are of a crisp type.

10. The apparatus of claim 9 wherein said evaluation signals each have a degree of fulfillment portion indicative of a degree of fulfillment for said condition portion of an associated one of said rules and a consequent portion indicative of said consequent portion of said associated rule, and wherein said synthesizer provides said estimate signal at a magnitude which is indicative of a degree of fulfillment weighted average of said consequent portion of said evaluation signals.

11. The apparatus of claim 10 wherein said difference signals are of representative of a crisp type and wherein said evaluator further comprises a fuzzifier responsive to said difference signals and having at least one set of fuzzy membership functions for providing fuzzy logic signals indicative of said difference signals to said fuzzy inferencer for use in determining said degree of fulfillment of said condition portions of said rules in said fuzzy rulebase.

12. A method for use in estimating an actual magnitude of a physical parameter, using at least three redundant input signals, each input signal is indicative of a sensed magnitude of the parameter, the method comprising:

providing in response to the at least three redundant input signals at least three difference signals indicative of the relative differences between the sensed magnitudes;

computing in response to said at least three difference signals, a plurality of evaluation signals; and

generating in response to said evaluation signals, an estimate signal indicative of an estimate of the actual magnitude of the physical parameter,

wherein said estimate signal has a magnitude indicative of a summation of the product of the magnitude indicated by each redundant input signal and an associated proportionality factor, wherein each proportionality factor has a value equal to one of a plurality of possible values for that proportionality factor, the value of each proportionality factor depending upon said magnitudes of said difference signals.

13. The method of claim 12 wherein said possible values of said proportionality factors ranges from substantially zero to substantially unity.

14. The method of claim 12 wherein said possible values of said proportionality factors comprises a generally continuous set of values.

15. The method of claim 12 wherein each of said difference signals has a magnitude proportional to the difference between the sensed magnitudes indicated by two of the redundant input signals.

16. The method of claim 12 wherein said estimate signal has a magnitude that is described as a smoothly varying function.

17. The method of claim 12 wherein said evaluation signals are of a type indicative of weighting factors and said estimate signal has a magnitude indicative of a weighted average based on said weighting factors.

18. The method as in claim 17 further comprising the step of providing a plurality of signals each indicative of a possible estimate of the actual magnitude of the physical parameter, and wherein said estimate signal is indicative of a weighted average of said plurality of signals.

19. A method for use in estimating an actual magnitude of a physical parameter at least three redundant input signals, each input signal is indicative of a sensed magnitude of the parameter, the method comprising:

providing in response to the at least three redundant input signals at least three difference signals indicative of the relative differences between the sensed magnitudes;

computing a plurality of evaluation signals using said at least three difference signals and the at least three redundant input signals,

wherein said computing step comprises a fuzzifying step for providing said evaluation signals as representative of fuzzy logic signals; and

generating in response to said evaluation signals, an estimate signal indicative of an estimate of the actual magnitude of the physical parameter.

20. The method of claim 19 wherein said computing step further comprises the step of fuzzy inferencing including

providing a fuzzy rulebase with a plurality of rules, each of said rules having a condition portion and a consequent portion, wherein said consequent portion is of a crisp type.

21. The method of claim 20 wherein said evaluation signals each have a degree of fulfillment portion indicative of a degree of fulfillment for said condition portion of an associated one of said rules and a consequent portion indicative of said consequent portion of said associated rule, and wherein said estimate signal has a magnitude which is indicative of a degree of fulfillment weighted average of said consequent portion of said evaluation signals.

22. The method of claim 19 wherein said difference signals are representative of a crisp type and wherein said step of computing further comprises the step of fuzzifying including

providing at least one set of fuzzy membership functions and evaluating said fuzzy membership functions with respect to said difference signals to produce representing said difference signals but of a type comprising fuzzy logic;

and wherein said step of fuzzy inferencing further comprises the step of determining in response to said fuzzy logic signals a degree of fulfillment for each of said condition portions of said rules in said fry rulebase.

23. An engine control system for use in operating a gas turbine engine, said system comprising:

a sensor array of at least three sensors for providing at least three redundant input signals corresponding to the value of an engine parameter; and

estimator apparatus having

means for providing, in response to the at least three redundant input signals at least three difference signals indicative of the relative differences between the respective sensor signal values;

means for computing a plurality of evaluation signals in response to said at least three difference signals; and

means for generating an estimate signal in response to said evaluation signals indicative of an estimate of the actual magnitude of the parameter,

wherein said estimate signal has a magnitude indicative of a summation of the product of the magnitude indicated by each redundant input signal and an associated proportionality factor, wherein each proportionality factor has a value equal to one of a plurality of possible values for that proportionality factor, the value of each proportionality factor depending upon said magnitudes of said difference signals.

24. The system of claim 23 wherein said possible values of said proportionality factors ranges from substantially zero to substantially unity.

25. The system of claim 23 wherein said possible values of said proportionality factors comprises a generally continuous set of values.

26. The system of claim 23 wherein each of said difference signals has a magnitude proportional to the difference between the sensed magnitudes indicated by two of the redundant input signals.

27. The system of claim 23 wherein said estimate signal has a magnitude that is described by a smoothly varying function.

28. The system as in claim 23 wherein said evaluation signals are of a type indicative of weighting factors and said estimate signal has a magnitude indicative of a weighted average based on said weighting factors.

29. The system as in claim 28 wherein said means for computing further receives the redundant input signals and provides in response to the redundant input signals a plurality of signals each indicative of a possible estimate of the actual magnitude of the physical parameter, and wherein said means for generating receives said plurality of signals and provides said estimate signal which is indicative of a weighted average of said plurality of signals.

34. The system of claim 23 further comprising a memory section wherein said estimate signals are stored in said memory section.

35. The system of claim 23 wherein one of said sensors comprises an analytical sensor presenting signals corresponding to a modeled value of said engine parameter.

36. The system of claim 35 wherein the sensed magnitude indicated by the input signal provided by the analytical sensor is not used in calculating the estimate unless there is disagreement between the sensed magnitudes indicated by the other input signals provided by the other sensors.

30. An engine control system for use in operating a gas turbine engine, said system comprising:

a sensor array of at least three sensors for providing at least three redundant input signals corresponding to the value of an engine parameter; and

estimator apparatus having

means for providing, in response to the at least three redundant input signals at least three difference signals indicative of the relative differences between the respective sensor signal values;

means for computing a plurality of evaluation signals in response to an input of said at least three difference signals and the at least three redundant input signals,

wherein said means for computing fuzzifies said evaluation signals as representative of fuzzy logic signals; and

means for generating an estimate signal in response to said evaluation signals indicative of an estimate of the actual magnitude of the parameter.

31. The system of claim 30 wherein said means for computing comprises a fuzzy inferencer having a rulebase with a plurality of rules, each of said rules having a condition portion and a consequent portion, and wherein said consequent portion are of a crisp type.

32. The system of claim 31 wherein said evaluation signals each have a degree of fulfillment portion indicative of a degree of fulfillment for said condition portion of an associated one of said rules and a consequent portion indicative of said consequent portion of said associated rule, and wherein said means for generating provides said estimate signal at a magnitude which is indicative of a degree of fulfillment weighted average of said consequent portion of said evaluation signals.

33. The system of claim 32 wherein said difference signals are of representative of a crisp type and wherein said means for computing further comprises a fuzzifier responsive to said difference signals and having at least one set of fuzzy membership functions for providing fuzzy logic signals indicative of said difference signals to said fuzzy inferencer for use in determining said degree of fulfillment of said condition portions of said rules in said fuzzy rulebase.