Methods and Apparatuses for Defect Diagnosis in a Mechanical System

• US 20150233792A1
• Filed: 05/31/2012
• Published: 08/20/2015
• Est. Priority Date: 05/31/2012
• Status: Active Grant

• Current Assignee: Canrig Drilling Technology Limited (Nabors Industries Limited), University of Connecticut (State of Connecticut)
• Original Assignee: Canrig Drilling Technology Limited (Nabors Industries Limited), University of Connecticut (State of Connecticut)
• Inventors: Gao, Robert X., Wang, Jinjiang, Yan, Ruqiang, Ellis, Brian Charles, Smith, Boone Elbert, Puente, Jose Abelardo Sanchez
• Granted Patent: US 10,520,397 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G01M 13/028 Acoustic or vibration analysis
• CPC Class Codes: G01M 13/045 Acoustic or vibration analysis
• CPC Class Codes: G01M 99/005 Testing of complete machines, e.g. washing-machines or mobile phones testing of machine parts G01M13/00; testing of electric apparatus or components G01R31/50
• CPC Class Codes: G01M 99/008 by doing functionality tests

Abstract

Multiple methods and corresponding apparatuses for efficient and reliable defect diagnosis in components of mechanical systems, are described. According to one aspect, multi-scale enveloping-order spectrogram is used to diagnose, or detect, defects in a moving component of a mechanical system. According to another aspect, defect identification and diagnosis in a motor is performed based on spectral characteristics of motor current envelope. According to yet another aspect, a logic rule model, employing classification of features associated with single- or multi-sensor data, is employed for diagnosis of defects in components of mechanical systems.

Claims

1. A method of detecting a defect in a mechanical system, the method comprising:

calculating a plurality of features using, at least in part, multi-sensor data associated with at least one component of the mechanical system;

classifying a subset of the calculated plurality of features based on classification patterns constructed from multi-sensor training data; and

determining whether a defect exists in the at least one component based on the classified subset of features.

2. The method of claim 1, wherein the mechanical system is a drilling rig.

3. The method of claim 1, wherein the calculated plurality of features include statistical features.

4. The method of claim 1, wherein the calculated plurality of features include frequency domain features.

5. The method of claim 4, wherein the frequency domain features include multi-scale enveloping order spectra features.

6. The method of claim 4, wherein the frequency domain features include motor current envelop spectral features.

7. The method of claim 1, wherein the calculated plurality of features include autoregressive coefficients.

8. The method of claim 1, wherein calculating a feature includes estimating the feature.

9. The method of claim 1 further comprising selecting a subset of the calculated plurality of features, the selected subset being classified.

10. The method of claim 9 wherein selecting a subset of the calculated plurality of features includes selecting a subset of the calculated plurality of features based on a minimum Redundancy Maximum Relevancy (mRMR) feature selection method.

11. The method of claim 9 wherein selecting a subset of the calculated plurality of features includes selecting a subset of the calculated plurality of features based on at least one of sequential backward selection method, principal component analysis method, and independent component analysis method.

12. The method of claim 1, wherein classifying the subset of features includes classifying the subset of features according to a support vector machine method.

13. The method of claim 1, wherein classifying the subset of features includes classifying the subset of features according to a k-nearest neighbor method.

14. The method of claim 1, wherein classifying the subset of features includes classifying the subset of features according to a Naï

ve Bayes method.

15. The method of claim 1, further comprising reporting whether a defect in the at least one component of the mechanical system exists.

16. The method of claim 15, wherein reporting whether a defect in the at least one component of the mechanical system exists includes initiating an alarm.

17. The method of claim 1, wherein the multi-sensor training data include measured data.

18. The method of claim 1, wherein the multi-sensor training data include computer simulated data.

37. An apparatus comprising means for performing the method of claim 1.

38. An apparatus comprising a memory with computer code instructions stored thereon and a processor, the memory and computer code instructions, with the processor, being configured to cause the apparatus to perform the method of claim 1.

39. A computer-readable medium comprising computer code instructions stored thereon, the computer code instructions when executed by a processor cause an apparatus to perform the method of claim 1.

19. A method of detecting a defect in a mechanical system, the method comprising:

calculating a set of wavelet envelopes of a time sampled first representation of a moving component of the mechanical system;

converting the set of wavelet envelopes from a time domain to a spatial domain based on measured samples of a time sampled second representation of motion associated with the moving component;

determining one or more spectral energy concentrations of the converted wavelet envelopes associated with at least one expected defect frequency of the moving component; and

determining whether a defect in the moving component exists based on the calculated one or more spectral energy concentrations.

20. The method of claim 19, wherein the mechanical system is a drilling rig.

21. The method of claim 19, wherein the first representation of motion is an acceleration signal, the second representation of motion is a speed signal, and the spatial domain is rotational angle domain.

22. The method of claim 19, wherein the second representation of motion is time-varying.

23. The method of claim 19, wherein the time sampled second representation of motion is sampled at a first set of sampling time instances with constant time period and wherein converting the set of wavelet envelopes includes:

calculating a second set of time instances corresponding to equal increments of spatial displacement; and

calculating another set of wavelet envelopes with samples corresponding to the second set of time instances.

24. The method of claim 19, wherein determining whether a defect in the moving component exists includes comparing the determined one or more spectral energy concentrations to one or more thresholds.

25. The method of claim 19, wherein determining whether a defect in the moving component exists includes comparing averages of the determined one or more spectral energy concentrations across different wavelet scales to one or more thresholds.

26. The method of claim 19, wherein determining whether a defect in the moving component exists includes comparing sums of the determined one or more spectral energy concentrations across different wavelet scales to one or more thresholds.

27. The method of claim 19, further comprising reporting whether a defect in the moving component exists.

28. The method of claim 27, wherein reporting whether a defect in the moving component exists includes initiating an alarm.

29. A method of detecting a defect in a mechanical system, the method comprising:

calculating an envelope of an electric current signal of a motor of the mechanical system;

calculating one or more spectral energy concentrations of the envelope at one or more expected defect frequencies of the motor; and

determining whether a defect in the motor exists based on the calculated one or more spectral energy concentrations.

30. The method of claim 29, wherein determining whether a defect in the motor exists includes comparing the determined one or more spectral energy concentrations to one or more thresholds.

31. The method of of claim 29 further including calculating other features of the envelope, other than the one or more spectral energy concentrations of the envelope at one or more expected defect frequencies of the motor, and wherein determining whether a defect in the motor exists includes classifying a subset of the calculated features, from the set of calculated features including the calculated one or more spectral energy concentrations and the other features, based on classifications patterns, the classification patterns being calculated using training data associated with the energy concentrations at expected defect frequencies of the motor and using training data associated with the other features.

32. The method of claim 31 further including selecting the subset of features based on a minimum redundancy maximum relevancy feature selection method.

33. The method of claim 29, further comprising reporting whether a defect in the motor exists.

34. The method of claim 33, wherein reporting whether a defect in the motor exists includes initiating an alarm.

35. The method of claim 33, wherein the motor is a drilling rig motor.

36. A method of detecting a defect in a mechanical system, the method comprising:

converting a set of measured samples of a time sampled first representation of motion associated with a moving component of the mechanical system from time domain to spatial domain based on measured samples of a time sampled second representation of motion associated with the moving component;

calculating a set of wavelet envelopes of the converted set of measured samples of the time sampled first representation of the moving component of the mechanical system;

determining one or more spectral energy concentrations of the calculated wavelet envelopes associated with at least one expected defect frequency of the moving components; and

determining whether a defect in the moving component exists based on the calculated one or more energy concentrations.