Probabilistic Stress Wave Analysis System and Method

• US 20070282545A1
• Filed: 05/17/2007
• Published: 12/06/2007
• Est. Priority Date: 05/17/2006
• Status: Active Grant

• Current Assignee: Curtiss-Wright Flow Control Corporation (Curtiss-Wright Corporation)
• Original Assignee: Curtiss-Wright Flow Control Corporation (Curtiss-Wright Corporation)
• Inventors: Board, David B.
• Granted Patent: US 7,505,852 B2
• Time in Patent Office: Days
• US Class Current: 702/56
• CPC Class Codes: G01H 1/00 Measuring characteristics of vibrations in solids by using direct conduction to the detector G01H9/00, G01H11/00 take precedence
• CPC Class Codes: G01M 15/12 by monitoring vibrations
• CPC Class Codes: G01M 5/0041 by determining deflection or stress
• CPC Class Codes: G01M 5/0066 by exciting or detecting vibration or acceleration vibration testing of structures G01M7/00
• 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: 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 29/4454 Signal recognition, e.g. specific values or portions, signal events, signatures
• CPC Class Codes: G05B 23/024 Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks

Abstract

A statistical process, and system for implementing the process, is described for the analysis of stress waves generated in operating machinery or equipment. This technique is called Probabilistic Stress Wave Analysis. The process is applied to a population of individual “feature” values extracted from a digitized time waveform (such as a 2 second Stress Wave Pulse Train, or a 2 month history of Stress Wave Energy). Certain numeric descriptors of the statistical distributions of computed features are then employed as inputs to decision making routines (such as neural networks or simple threshold testing) to accurately classify the condition represented by the original time waveform data, and thereby determine a status of the operating machine/equipment.

Claims

1. A method for determining a current status of an apparatus, said method comprising the steps of:

monitoring stress waves generated by the apparatus during a normal operation of the apparatus for at least one time period to generate first stress wave data;

analyzing at least a portion of said first stress wave data to generate first analyzed data that captures one or more features of said first stress wave data;

processing at least a portion of said first analyzed data to generate baseline data, wherein said baseline data has substantially reduced influence from normal operating variations compared to said first analyzed data;

further monitoring stress waves generated by the apparatus during a current operation of the apparatus for a current time period to generate second stress wave data;

analyzing at least a portion of said second stress wave data to generate second analyzed data that captures one or more features of said second stress wave data;

processing at least a portion of said second analyzed data to generate current status data, wherein said current status data also has substantially reduced influence from normal operating variations compared to said second analyzed data; and

comparing said current status data to said baseline data to determine the current status of the apparatus.

2. The method of claim 1, wherein said baseline data includes baseline probabilistic data representing a normal operating condition of the apparatus, and wherein said current status data includes current probabilistic data representing a current operating condition, and further wherein said comparing step compares said baseline probabilistic data to said current probabilistic data to determine said current status of the apparatus.

3. The method of claim 2, wherein both of said probabilistic data includes statistical measures of the Probability Density Function of the corresponding source data.

4. The method of claim 2, wherein both of said probabilistic data include the results of one or more of:

a third moment test for a normal distribution, a maximum value of the population, a ratio of (the maximum−

the mean) divided by (the maximum−

the minimum), a ratio of the standard deviation of the population to the mean of the population, a skewness coefficient, and the kurtosis of the population.

5. The method of claim 4, wherein said comparing step examines a distortion of the probabilistic data of said current status data as compared to the probabilistic data of said baseline data to determine if a failure or a deterioration of the apparatus currently exists, or is likely to occur at some point in the future.

6. The method of claim 1, wherein said first stress wave data includes data obtained from a first location of the apparatus, and also includes data obtained from a second location of the apparatus different from said first location.

7. The method of claim 6, wherein said baseline data is derived from a difference between the portion of said first analyzed data obtained from said first location and the portion of said first analyzed data obtained from said second location.

8. The method of claim 6, wherein said second stress wave data includes data obtained from said first location of the apparatus, and also includes data obtained from said second location of the apparatus.

9. The method of claim 8, wherein

said baseline data is derived from a difference between the portion of said first analyzed data obtained from said first location and the portion of said first analyzed data obtained from said second location, and further wherein

said current status data is derived from a difference between the portion of said second analyzed data obtained from said first location and the portion of said second analyzed data obtained from said second location.

10. The method of claim 9, wherein said baseline data provides a normal operating threshold of the apparatus, and wherein, when said comparing of said current status data to said baseline data results in said current status data falling outside of said normal operating threshold, it is determined that a failure or a deterioration of the apparatus currently exists, or is likely to occur at some point in the future.

11. A method for determining a current status of an apparatus, said method comprising the steps of:

monitoring stress waves generated by the apparatus during a current operation of the apparatus for a current time period to generate current stress wave data;

analyzing at least a portion of said current stress wave data to generate current analyzed data that captures one or more features of said current stress wave data;

processing at least a portion of said current analyzed data to generate probabilistic current status data, wherein said probabilistic current status data has substantially reduced influence from normal operating variations compared to said current analyzed data; and

comparing said probabilistic current status data to probabilistic baseline data to determine the current status of the apparatus.

12. The method of claim 11, wherein both of said probabilistic data includes statistical measures of the Probability Density Function of the corresponding source data.

13. The method of claim 11, further comprising the step of determining said probabilistic baseline data by a method including the steps of:

monitoring stress waves generated by the apparatus during a normal operation of the apparatus for a plurality of time periods to generate a baseline series of stress wave data;

analyzing at least a portion of said baseline series of stress wave data to generate a baseline series of analyzed data that captures one or more features of said baseline series of stress wave data; and

processing at least a portion of said baseline series of analyzed data to generate said probabilistic baseline data.

14. The method of claim 13, wherein both of said probabilistic baseline data and said probabilistic current status data include the results of one or more of:

a third moment test for a normal distribution, a maximum value of the population, a ratio of (the maximum−

the mean) divided by (the maximum−

the minimum), a ratio of the standard deviation of the population to the mean of the population, a skewness coefficient, and the kurtosis of the population.

15. The method of claim 14, wherein said comparing step examines a distortion of the said probabilistic current status data as compared to said probabilistic baseline data to determine if a failure or a deterioration of the apparatus currently exists, or is likely to occur at some point in the future.

16. The method of claim 11, wherein said probabilistic current status data include the results of one or more of:

a third moment test for a normal distribution, a maximum value of the population, a ratio of (the maximum−

the mean) divided by (the maximum−

the minimum), a ratio of the standard deviation of the population to the mean of the population, a skewness coefficient, and the kurtosis of the population.

17. The method of claim 11, wherein said comparing step examines a distortion of the said probabilistic current status data as compared to said probabilistic baseline data to determine if a failure or a deterioration of the apparatus currently exists, or is likely to occur at some point in the future.

18. A method for determining a current status of an apparatus, said method comprising the steps of:

monitoring stress waves generated by the apparatus at a first location during a current operation of the apparatus for a current time period to generate first current stress wave data;

analyzing at least a portion of said first current stress wave data to generate first current analyzed data that captures one or more features of said first current stress wave data;

monitoring stress waves generated by the apparatus at a second location different than said first location during the current operation of the apparatus for said current time period to generate second current stress wave data;

analyzing at least a portion of said second current stress wave data to generate second current analyzed data that captures one or more features of said current second stress wave data;

processing at least a portion of said first current analyzed data and said second current analyzed data to generate current status data based on a difference between said first current analyzed data and said second current analyzed data; and

comparing said current status data to a threshold to determine the current status of the apparatus.

19. The method of claim 18, further comprising the step of determining said threshold by a method including the steps of:

monitoring stress waves generated by the apparatus at said first location during a normal operation of the apparatus for at least one time period to generate first initial stress wave data;

analyzing at least a portion of said first initial stress wave data to generate first initial analyzed data that captures one or more features of said first initial stress wave data;

monitoring stress waves generated by the apparatus at said second location during the normal operation of the apparatus for said at least one time period to generate second initial stress wave data;

analyzing at least a portion of said second initial stress wave data to generate second initial analyzed data that captures one or more features of said initial second stress wave data; and

processing at least a portion of said first initial analyzed data and said second initial analyzed data to generate said threshold.

20. The method of claim 19, wherein said threshold is based on a difference between said first initial analyzed data and said second initial analyzed data.

21. The method of claim 19, said comparing is for determining if a failure or a deterioration of the apparatus currently exists, or is likely to occur at some point in the future.

22. The method of claim 18, wherein said threshold includes baseline PDF descriptors, and wherein said current status data includes current PDF descriptors such that said comparing includes comparing said current PDF descriptors to said baseline PDF descriptors.

23. A system for determining a current status of an apparatus, said system comprising:

at least one sensor mounted on or in the apparatus for monitoring stress waves generated by the apparatus during a normal operation of the apparatus for at least one time period to generate first stress wave data;

a feature extraction device adapted for analyzing at least a portion of said first stress wave data to generate first analyzed data that captures one or more features of said first stress wave data;

means for processing at least a portion of said first analyzed data to generate baseline data, wherein said baseline data has substantially reduced influence from normal operating variations compared to said first analyzed data;

means for further monitoring stress waves generated by the apparatus during a current operation of the apparatus for a current time period to generate second stress wave data;

means for analyzing at least a portion of said second stress wave data to generate second analyzed data that captures one or more features of said second stress wave data;

means for processing at least a portion of said second analyzed data to generate current status data, wherein said current status data also has substantially reduced influence from normal operating variations compared to said second analyzed data; and

means comparing said current status data to said baseline data to determine the current status of the apparatus.

24. The system of claim 23, wherein said baseline data includes baseline probabilistic data representing a normal operating condition of the apparatus, and wherein said current status data includes current probabilistic data representing a current operating condition, and further wherein said comparing step compares said baseline probabilistic data to said current probabilistic data to determine said current status of the apparatus.

25. The system of claim 24, wherein both of said probabilistic data includes statistical measures of the Probability Density Function of the corresponding source data.

26. The system of claim 24, wherein both of said probabilistic data include the results of one or more of:

a third moment test for a normal distribution, a maximum value of the population, a ratio of (the maximum−

the mean) divided by (the maximum−

the minimum), a ratio of the standard deviation of the population to the mean of the population, a skewness coefficient, and the kurtosis of the population.

27. The system of claim 26, wherein said comparing step examines a distortion of the probabilistic data of said current status data as compared to the probabilistic data of said baseline data to determine if a failure or a deterioration of the apparatus currently exists, or is likely to occur at some point in the future.

28. The system of claim 23, said system further comprising at least one other sensor mounted in or on the apparatus at another location, wherein said first stress wave data includes data obtained from said at least one sensor and also data obtained from said at least one other sensor.

29. The system of claim 28, wherein said baseline data is derived from a difference between the data obtained from said at least one sensor and the data obtained from said at least one other sensor.

30. The system of claim 28, wherein said second stress wave data includes data obtained from said at least one sensor, and also includes data obtained from said at least one other sensor.

31. The system of claim 30, wherein

said baseline data is derived from a difference between the portion of said first analyzed data obtained from said at least one sensor and the portion of said first analyzed data obtained from said at least one other sensor, and further wherein

said current status data is derived from a difference between the portion of said second analyzed data obtained from said at least one sensor and the portion of said second analyzed data obtained from said at least one other sensor.

32. The system of claim 31 wherein said baseline data provides a normal operating threshold of the apparatus, and wherein, when said comparing of said current status data to said baseline data results in said current status data falling outside of said normal operating threshold, said system determines that a failure or a deterioration of the apparatus currently exists, or is likely to occur at some point in the future.

33. A system for determining a current status of an apparatus, said system comprising:

a sensor mounted on or in the apparatus for monitoring stress waves generated by the apparatus during a current operation of the apparatus for a current time period to generate current stress wave data;

means for analyzing at least a portion of said current stress wave data to generate current analyzed data that captures one or more features of said current stress wave data;

means for processing at least a portion of said current analyzed data to generate probabilistic current status data, wherein said probabilistic current status data has substantially reduced influence from normal operating variations compared to said current analyzed data; and

means for comparing said probabilistic current status data to probabilistic baseline data to determine the current status of the apparatus.

34. A system for determining a current status of an apparatus, said system comprising:

a first sensor mounted on or in the apparatus at a first location for monitoring stress waves generated by the apparatus during a current operation of the apparatus for a current time period to generate first current stress wave data;

means for analyzing at least a portion of said first current stress wave data to generate first current analyzed data that captures one or more features of said first current stress wave data;

a second sensor mounted on or in the apparatus at a second location different than said first location for monitoring stress waves generated by the apparatus during the current operation of the apparatus for said current time period to generate second current stress wave data;

means for analyzing at least a portion of said second current stress wave data to generate second current analyzed data that captures one or more features of said current second stress wave data;

means for processing at least a portion of said first current analyzed data and said second current analyzed data to generate current status data based on a difference between said first current analyzed data and said second current analyzed data; and

means for comparing said current status data to a threshold to determine the current status of the apparatus.

35. The system of claim 34, wherein said threshold includes baseline PDF descriptors, and wherein said current status data includes current PDF descriptors such that said means for comparing includes means for comparing said current PDF descriptors to said baseline PDF descriptors.