Method and apparatus for determining the health of a component using condition indicators

• US 6,728,658 B1
• Filed: 12/04/2001
• Issued: 04/27/2004
• Est. Priority Date: 05/24/2001
• Status: Expired due to Term

• Current Assignee: Simmonds Precision Products, Inc. (Raytheon Technologies Corporation d/b/a RTX)
• Original Assignee: Simmonds Precision Products, Inc. (Raytheon Technologies Corporation d/b/a RTX)
• Inventors: Bechhoefer, Eric Robert
• Primary Examiner(s): Wachsman, Hal
• Application Number:
  US10/011,787
• Time in Patent Office: 875 Days
• Field of Search: 702/181-185, 702/33-36, 702/56, 702/81, 702/113-115, 702/117, 702/119, 702/187, 702/188, 702/FOR-123, 702/FOR-134, 702/FOR.139, 702/FOR.426, 702/FOR.171, 702/FOR.172, 706/904-906, 706/916, 706/912-914, 703/6-8, 700/51, 700/109, 700/110, 700/175, 701/29, 701/30, 73/350.9, 735/70, 738/02, 340/674, 340/680, 340/682, 340/683
• US Class Current: 702/181
• CPC Class Codes: G01H 1/003 of rotating machines G01H1/10 takes precedence
• CPC Class Codes: G01N 2291/2693 Rotor or turbine parts
• CPC Class Codes: G01N 2291/2696 Wheels, Gears, Bearings
• CPC Class Codes: G01N 29/11 by measuring attenuation of acoustic waves
• CPC Class Codes: G01N 29/12 by measuring frequency or resonance of acoustic waves measuring frequency or resonant frequency of mechanical vibrations or acoustic waves in general G01H1/06, G01H3/04, G01H13/00; acoustic resonators G10K11/04; vibration or shock testing of structures G01M7/00
• CPC Class Codes: G01N 29/4445 Classification of defects
• CPC Class Codes: G01N 29/449 Statistical methods not provided for in G01N29/4409, e.g. averaging, smoothing and interpolation
• CPC Class Codes: G01N 29/46 by spectral analysis, e.g. Fourier analysis or wavelet analysis spectral signal processing per se G06F17/14
• CPC Class Codes: G05B 23/0235 based on a comparison with predetermined threshold or range, e.g. "classical methods", carried out during normal operation; threshold adaptation or choice; when or how to compare with the threshold

Abstract

Disclosed are techniques used in connection with determining a health indicator (HI) of a component, such as that of an aircraft component. The HI is determined using condition indicators (CIs) which parameterize characteristics about a component minimizing possibility of a false alarm. Different algorithms are disclosed which may be used in determining one or more CIs. The HI may be determined using a normalized CI value. Techniques are also described in connection with selecting particular CIs that provide for maximizing separation between HI classifications. Given a particular HI at a point in time for a component, techniques are described for predicting a future state or health of the component using the Kalman filter. Techniques are described for estimating data values as an alternative to performing data acquisitions, as may be used when there is no pre-existing data.

Claims

1. A method for determining a health indicator associated with a component, comprising:

determining a plurality of health classifications;

determining at least one condition indicator guantifying a characteristic of said component;

determining a probability associated with each of said health classifications, said probability being an estimation that said component is of a particular health classification given said at least one condition indicator; and

determining, for a given set of observed values, which of said plurality of health classifications is associated with said component using said probabilities associated with said health classifications,

wherein said determining at least one condition indicator, said determining a probability, and said determining which of said health classifications is associated with said component, are performed upon two of said plurality of health classifications, and

wherein, if a determination is made that a first of said health classifications is not associated with said component, repeating said determining at least one condition indicator, said determining a probability, and said determining which of said health classifications is associated with said component using two other health classifications.

2. The method of claim 1, further comprising:

storing previously observed and calculated data associated with said component; and

using a portion of said previously observed and calculated data to determine said probability associated with said each health classification.

3. The method of claim 1, further comprising:

obtaining data estimates using models of said component; and

using said data estimates to determine said probability associated with said each health classification.

4. A method for determining a health indicator associated with a component, comprising:

determining a plurality of health classifications;

determining at least one condition indicator quantifying a characteristic of said component;

determining a probability associated with each of said health classifications, said probability being an estimation that said component is of a particular health classification given said at least one condition indicator;

determining, for a given set of observed values, which of said plurality of health classifications is associated with said component using said probabilities associated with said health classifications;

determining a threshold using a ratio of a portion of said probabilities;

calculating a current quantity associated with said component using said observed values; and

comparing said current quantity to said ratio to determine which of said health classifications is associated with said component.

5. The method of claim 4, wherein said calculating said current quantity further comprises:

determining a transformation matrix maximizing a distance between two of said plurality of health classifications;

determining a covariance matrix for each of said health classifications using said at least one condition indicator and said observed values; and

determining said current quantity in accordance with said covariance matrices, said transformation matrix and said observed values.

6. The method of claim 5, wherein, for two of said plurality of health classifications, said current quantity, h(X), may be represented as:

7. A method for determining a health status of a component comprising:

selecting a plurality of condition indicators having a value and each having a corresponding weighting factor, and at least one threshold value defining at least two health classifications;

determining a contribution to a health indicator for each of said condition indicators, wherein said determining further comprises, for each of said plurality of indicators;

determining which of said at least two classifications said value of said each condition indicator belongs; and

determining said contribution to said health indicator by said each condition indicator in accordance with a selected one of said at least two classifications and said weighted factor; and

determining said health indicator according to all contributions by each of said condition indicator values.

8. The method of claim 7, wherein first, second and third classifications are associated with each of said condition indicators formed by two threshold values, and the method further comprises:

determining that said contribution to said health indicator for a first of said condition indicators is zero if said value of said condition indicator is in said first classification;

determining that said contribution to said health indicator is a first multiple of said weighting factor if said value of said condition indicator is in said second classification; and

determining that said contribution to said health indicator is a second multiple of said weighting factor if said value of said condition indicator is in said third classification.

9. The method of claim 8, where the two threshold values are alarm level and warning level.

10. A computer program product executed on a processor for determining a health indicator associated with a component, comprising machine executable code for

determining a plurality of health classifications;

determining at least one condition indicator quantifying a characteristic of said component;

determining a probability associated with each of said health classifications, said probability being an estimation that said component is of a particular health classification given said at least one condition indicator; and

determining, for a given set of observed values, which of said plurality of health classifications is associated with said component using said probabilities associated with said health classifications,

wherein machine executable code for determining at least one condition indicator, determining a probability, and determining which of said health classifications is associated with said component, are performed upon two of said plurality of health classifications, and the computer program product further includes machine executable code, wherein, if a determination is made that a first of said health classifications is not associated with said component, for repeatedly determining at least one condition indicator, repeatedly determining a probability, and repeatedly determining which of said health classifications is associated with said component using two other health classifications.

11. The computer program product of claim 10, further comprising machine executable code for:

storing previously observed and calculated data associated with said component; and

using a portion of said previously observed and calculated data to determine said probability associated with each health classification.

12. The computer program product of claim 10, further comprising machine executable code for:

obtaining data estimates using models of said component; and

using said data estimates to determine said probability associated with said each health classification.

13. A computer program product executed on a processor for determining a health indicator associated with a component, comprising machine executable code for

determining a plurality of health classifications;

determining at least one condition indicator quantifying a characteristic of said component;

determining a probability associated with each of said health classifications, said probability being an estimation that said component is of a particular health classification given said at least one condition indicator;

determining, for a given set of observed values, which of said plurality of health classifications is associated with said component using said probabilities associated with said health classifications,

determining a threshold using a ratio of a portion of said probabilities;

calculating a current quantity associated with said component using said observed values; and

comparing said current quantity to said ratio to determine which of said health classifications is associated with said component.

14. The computer program product of claim 13, wherein said machine executable code for calculating said current quantity further comprises machine executable code for:

determining a transformation matrix maximizing a distance between two of said plurality of health classifications;

determining a covariance matrix for each of said health classifications using said at least one condition indicator and said observed values; and

determining said current quantity in accordance with said covariance matrices, said transformation matrix and said observed values.

15. The computer program product of claim 14, wherein, for two of said plurality of health classifications, said current quantity, h(X), may be represented as:

16. A computer program product executed on a processor for determining a health status of a component comprising machine executable code for:

selecting a plurality of condition indicators having a value and each having a corresponding weighting factor, and at least one threshold value defining at least two health classifications;

determining a contribution to a health indicator for each of said condition indicators, wherein said machine executable code for determining further comprises, for each of said plurality of indicators, machine executable code for;

determining which of said at least two classifications said value of said each condition indicator belongs; and

determining said contribution to said health indicator by said each condition indicator in accordance with a selected one of said at least two classifications and said weighted factor; and

determining said health indicator according to all contributions by each of said condition indicator values.

17. The computer program product of claim 16, wherein first second and third classification are associated with each of said condition indicators formed by two threshold values, and the computer program product further comprises machine executable code for:

determining that said contribution to said health indicator for a first of said condition indicators is zero if said condition indicator is in said first classification;

determining that said contribution to said health indicator is a first multiple of said weighting factor if said condition indicator is in said second classification; and

determining that said contribution to said health indicator is a second multiple of said weighting factor if said condition indicator is in said third classification.

18. The computer program product of claim 17, where the two threshold values are alarm level and warning level.