Harmonic activity locator

• US 20030130810A1
• Filed: 12/04/2002
• Published: 07/10/2003
• Est. Priority Date: 12/04/2001
• Status: Active Grant

• Current Assignee: SKF USA Incorporated (SKF)
• Original Assignee: SKF Condition Monitoring Inc
• Inventors: Smulders, Adrianus J., Wei, Jim J., Boerhout, Johannes I.
• Granted Patent: US 6,792,360 B2
• Time in Patent Office: Days
• US Class Current: 702/56
• CPC Class Codes: G01H 1/003 of rotating machines G01H1/10 takes precedence
• CPC Class Codes: G01M 13/028 Acoustic or vibration analysis
• CPC Class Codes: G01M 7/02 Vibration-testing by means of a shake table
• 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 29/14 using acoustic emission techniques echo of particles G01N29/046; measuring mechanical vibrations or acoustic waves in solids in general G01H1/00
• CPC Class Codes: G01N 29/46 by spectral analysis, e.g. Fourier analysis or wavelet analysis spectral signal processing per se G06F17/14

Abstract

Systems and methods for identifying the presence of a defect in vibrating machinery. An exemplary method comprises analysis of frequency spectrum vibration data of the machine. The method comprises deriving a harmonic activity index based on estimates of the energy associated with the frequency spectrum and the energy associated with the defect'"'"'s harmonic series. The method may comprise deriving a value K by estimating a value M indicative of the energy of the defect'"'"'s harmonic series and dividing M by the number of spectral lines corresponding to the defect'"'"'s harmonic series. The method may further comprise deriving a value R by estimating a value Q indicative of the energy in the frequency spectrum data and dividing Q by the number of spectral lines of the frequency spectrum data. The method further comprises deriving the harmonic activity index based on the estimated K and R. Related systems for executing the methods are also disclosed.

Claims

1. A method of identifying a component defect in a machine subject to vibrations, the method comprising:

estimating from frequency domain vibration data a value R indicative of the spectral energy of said vibrations;

estimating from said data a value K indicative of the spectral energy associated with said component defect; and

deriving a harmonic activity index based at least in part on the estimated values K and R.

2. The method of claim 1, wherein estimating R comprises evaluating the total number of spectral lines of said frequency domain vibration data.

3. The method of claim 1, wherein estimating K comprises adding amplitudes of a plurality of spectral lines that are harmonically related.

4. The method of claim 2, wherein estimating R comprises dividing a sum of all the amplitudes of said spectral lines by said total number of spectral lines.

5. The method of claim 4, wherein estimating K comprises adding amplitudes of a plurality of spectral lines that are harmonically related, and wherein estimating K further comprises dividing the result of said adding by the number of spectral lines in said plurality of spectral lines that are harmonically related.

6. The method of claim 5, wherein deriving said index comprises evaluating a ratio that is based at least in part on dividing K by R.

7. The method of claim 5, wherein deriving said index comprises evaluating a ratio that is based at least in part on dividing K by the difference between R and K.

8. The method of claim 6, further comprising issuing a warning if the index is greater than 2.

9. The method of claim 1, further comprising determining whether the index indicates a component defect condition.

10. A method of differentiating between vibration measurements indicative of the presence of a component defect in a machine and vibration measurements unrelated to the component defect, the method comprising:

receiving a frequency spectrum associated with said machine;

estimating a component defect fundamental frequency and harmonic frequencies and associated amplitudes;

estimating a value K indicative of the total energy associated with said fundamental and harmonic frequencies;

estimating a value R indicative of the total energy associated with said spectrum;

deriving a value I_HAL based at least in part on the estimated values K and R; and

determining based at least in part on I_HAL and the fundamental frequency of the component defect whether the vibrations of the machine are produced by the component defect.

11. The method of claim 10, wherein receiving a frequency spectrum comprises receiving data indicative of amplitudes and frequencies of vibrations of said machine.

12. The method of claim 10, wherein estimating a component defect fundamental frequency comprises interpolating values associated with contiguous spectral lines.

13. The method of claim 12, wherein interpolating values comprises selecting a triplet of spectral lines, wherein an amplitude of a middle line of said triplet of spectral lines is greater than an amplitude of other spectral lines of said triplet of spectral lines.

14. The method of claim 10, wherein estimating a value K comprises obtaining a sum M comprising the sum of all of the amplitudes associated with the fundamental and harmonic frequencies of said component defect.

15. The method of claim 14, wherein estimating a value K further comprises dividing M by the number of harmonics N, including the fundamental frequency, associated with the component defect.

16. The method of claim 10, wherein estimating a value R comprises obtaining a sum Q comprising the sum of all of the amplitudes corresponding to the spectral lines in said spectrum.

17. The method of claim 16, wherein estimating a value R further comprises dividing Q by the total number of spectral lines P in said spectrum.

18. The method of claim 17, wherein deriving I_HAL comprises dividing K by R.

19. The method of claim 17, wherein deriving I_HAL comprises dividing K by the difference between K and R.

20. The method of claim 10, wherein deriving I_HAL comprises dividing K by R.

21. The method of claim 10, wherein deriving I_HAL comprises dividing K by the difference between K and R.

22. A system for identifying a component defect in a machine subject to vibrations, the system comprising:

a data storage module that receives and stores data indicative of amplitudes of vibrations of said machine at selected frequencies;

a data analyzer module, in communication with said data storage module, that derives a harmonic activity index, wherein said data analyzer comprises computer instructions operative for;

estimating from said data a value R indicative of the spectral energy of said vibrations;

estimating from said data a value K indicative of the spectral energy associated with said component defect; and

deriving said harmonic activity index based at least in part on the estimated values K and R.

23. The system of claim 22, wherein said computer instructions for deriving said index comprises computer instructions for dividing K by R.

24. The system of claim 22, wherein said computer instructions for deriving said index comprises computer instructions for dividing K by a difference between R and K.

25. The system of claim 22, wherein estimating K is based at least in part on adding a plurality of amplitudes corresponding to a harmonic series of a fundamental frequency of said component defect.

26. A system for identifying a component defect in a machine subject to vibrations, the method comprising:

means for receiving data indicative of amplitudes and corresponding frequencies of vibrations of said machine;

means for estimating from said data a value R indicative of the spectral energy of said vibrations;

means for estimating from said data a value K indicative of the spectral energy associated with said component defect, wherein estimating K is based at least in part on adding a plurality of amplitudes corresponding to a harmonic series of the fundamental frequency of said component defect; and

means for deriving a harmonic activity index based at least in part on the estimated values K and R.

27. A method of evaluating a frequency domain spectrum of vibration data comprising:

defining a first fundamental frequency;

defining a series of harmonics of said first fundamental frequency;

summing a first set of amplitudes associated with said first fundamental frequency and at least some of said harmonics to produce a value K;

summing a second set of amplitudes to produce a value R;

calculating a value based at least in part on said first and second sums.

28. The method of claim 27, further comprising comparing said value to a threshold.

29. The method of claim 27, wherein defining a first fundamental frequency comprises interpolating an amplitude between a pair of spectral lines of said frequency domain spectrum.

30. The method of claim 27, wherein defining a first fundamental frequency comprises searching for a spectral line triplet within a search band in said frequency domain spectrum, said spectral line triplet having a central spectral line amplitude larger than the amplitude of either spectral line adjacent to said central spectral line amplitude.

31. The method of claim 27, further comprising defining a second fundamental frequency, and repeating said harmonic defining, summing, and calculating acts using said second fundamental frequency.

32. The method of claim 31, further comprising repeatedly defining a series of additional fundamental frequencies and repeating said harmonic defining, summing, and calculating acts using each of said additional fundamental frequencies so as to produce a corresponding series of values.

33. The method of claim 32, wherein defining each of said fundamental frequencies comprises searching for a spectral line triplet within at least one search band in said frequency domain spectrum, said spectral line triplet having a central spectral line amplitude larger than the amplitude of either spectral line adjacent to said central spectral line amplitude.

34. The method of claim 33, further comprising searching within a plurality of search bands.

35. The method of claim 34, further comprising performing an exhaustive search throughout said frequency domain spectrum for said spectral line triplets.