Extensions to dynamically configurable process for diagnosing faults in rotating machines

• US 6,980,910 B1
• Filed: 03/25/2003
• Issued: 12/27/2005
• Est. Priority Date: 03/20/2001
• Status: Expired due to Fees

• Current Assignee: Johnson Controls Technology Company (Johnson Controls International plc)
• Original Assignee: Johnson Controls Technology Company (Johnson Controls International plc)
• Inventors: Verhaalen, Rick A., Shen, Hwa N.
• Primary Examiner(s): Bui, Bryan
• Application Number:
  US10/396,089
• Time in Patent Office: 1,008 Days
• Field of Search: 702/33, 702/34, 702/35, 702/56, 702/182, 702/183, 702/185, 702/189, 702/190, 734/62, 735/79, 735/59, 736/02, 736/60
• US Class Current: 702/56
• CPC Class Codes: G01H 1/003 of rotating machines G01H1/10 takes precedence
• CPC Class Codes: G01M 15/12 by monitoring vibrations
• CPC Class Codes: G01R 31/343 in operation

Abstract

A method and apparatus for detecting problems in a rotating machine is disclosed. The method and apparatus include creating a definition of the rotating machine using a plurality of templates that specify components and functional characteristics of the components. In addition, the method and apparatus include identifying a plurality of sensors configured to provide operational data related to the rotating machine and defining a plurality of problem rules including logical expressions by which machine faults are deduced from specified portions of the operational data. Further, the method and apparatus include performing an analysis of operation of the rotating machine, the analysis including acquiring actual operational data from the plurality of sensors and applying the problem rules and templates to the operational data to detect faults in the rotating machine and in response thereto generating an indication when a fault is detected.

Claims

1. A method for diagnosing faults in a rotating machine, said method comprising:

creating a definition of the rotating machine using a plurality of templates that specify components and functional characteristics of the components;

identifying a plurality of sensors configured to provide actual operational data related to the rotating machine;

defining a plurality of problem rules including logical expressions by which machine faults are deduced from specified portions of the actual operational data; and

performing an analysis of operation of the rotating machine, the analysis including acquiring the actual operational data from the plurality of sensors and applying the problem rules and templates to the actual operational data to detect faults in the rotating machine and in response thereto generating an indication when a fault is detected.

2. The method of claim 1, wherein the creating step comprises:

selecting at least one base template from a library of pre-existing base templates; and

modifying the at least one base template with data defining operational characteristics of the components.

3. The method of claim 1, wherein acquiring the actual operational data from the plurality of sensors includes acquiring data indicating vibration of the rotating machine.

4. The method of claim 1, wherein acquiring the actual operational data from the plurality of sensors includes obtaining vibration spectra.

5. The method of claim 4, wherein applying the problem rules and templates to the actual operational data includes searching an expected frequency range in the vibration spectra.

6. The method of claim 5, wherein applying the problem rules and templates to the actual operational data further includes extracting a peak amplitude and corresponding fundamental frequency from within the expected frequency range.

7. The method of claim 1, wherein the creating step includes using a generic single value template for defining variables of interest in the analysis.

8. The method of claim 1, wherein the creating step includes using a generic single value equation template to define frequency parameters over which variables of interest may be identified.

9. The method of claim 1, wherein the creating step includes using a base template for modeling characteristics of a component in an unassembled state.

10. The method of claim 1, wherein the creating step includes using a component template for modeling characteristics of a component in a power train.

11. The method of claim 1, wherein the creating step includes using at least one of the following templates:

a generic single value for defining a variable of interest in the analysis;

a generic single value equation to define frequency parameters over which variables of interest may be identified;

a base template for modeling characteristics of a component in an unassembled state; and

a component template for modeling characteristics of a component in a power train.

12. A method for diagnosing faults in a rotating machine, said method comprising:

creating templates that provide a definition of the rotating machine, the templates representing components and functional characteristics of the components; and

performing an analysis of operation of the rotating machine, the analysis including acquiring actual operational data from a plurality of sensors associated with the rotating machine, and applying the templates to the actual operational data using problem rules to detect faults in the rotating machine and in response thereto generating an indication when a fault is detected.

13. The method of claim 12, wherein applying the templates to the actual operational data includes using problem rules incorporating at least one of the following templates:

a generic single value for defining a variable of interest in the analysis;

a generic single value equation to define frequency parameters over which variables of interest may be identified;

a base template for modeling characteristics of a component in an unassembled state; and

a component template for modeling characteristics of a component in a power train.

14. The method of claim 12, wherein applying the templates to the actual operational data includes using problem rules incorporating the following templates:

a generic single value for defining a variable of interest in the analysis;

a generic single value equation to define frequency parameters over which variables of interest may be identified;

a base template for modeling characteristics of a component in an unassembled state; and

a component template for modeling characteristics of a component in a power train.

15. The method of claim 12, wherein creating the templates comprises using the definition of the rotating machine and formulae by which machine faults are deduced from specified portions of the actual operational data.

16. The method of claim 12, wherein acquiring actual operational data from the plurality of sensors includes obtaining vibration spectra.

17. A method of defining rules that specify fault conditions and severity in a rotating machine, the method comprising:

creating rules based on components and operational characteristics of the components, wherein the rules define formulae by which the rotating machine faults are deduced from operational data; and

utilizing a hierarchical tree-based structure to arrange the rules, the structure allowing an arbitrary grouping of logical expressions.

18. The method of claim 17, wherein the rules include at least one of the following templates:

a generic single value for defining a variable of interest in the analysis;

a generic single value equation to define frequency parameters over which variables of interest may be identified;

a base template for modeling characteristics of a component in an unassembled state; and

a component template for modeling characteristics of a component in a power train.

19. The method of claim 18, where the hierarchical tree-based structure includes nodes that define logical expressions or groups of logical expressions.

20. The method of claim 19, where a node specifies a group of logical expressions and an individual logical expression contained in the group is further contained in a subtree.

21. A method for diagnosing faults in a rotating machine, said method comprising:

defining rules based on components and operational characteristics of the components, where the rules define formulae by which rotating machine faults are deduced from actual operational data;

utilizing a hierarchical tree-based structure to arrange the rules, the structure allowing an arbitrary grouping of logical expressions; and

performing an analysis of operation of the rotating machine, the analysis including acquiring actual operational data from a plurality of sensors associated with the rotating machine, and applying the rules to the actual operational data to detect faults in the rotating machine and in response thereto generating an indication when a fault is detected.

22. The method of claim 21, wherein defining rules comprises creating templates including at least one of the following templates:

a generic single value for defining a variable of interest in the analysis;

a generic single value equation to define frequency parameters over which variables of interest may be identified;

a base template for modeling characteristics of a component in an unassembled state; and

a component template for modeling characteristics of a component in a power train.

23. A method for analyzing operational data in a rotating machine, said method comprising:

creating a first plurality of templates for defining the rotating machine;

creating a second plurality of templates for defining parameters of interest and search criteria; and

performing an analysis of operation of the rotating machine, the analysis including acquiring actual operational data from a plurality of sensors associated with the rotating machine, and applying the first and second plurality of templates to the actual operational data in the rotating machine.

24. The method of claim 23, wherein a power train approach is used to traverse components of the machine and find the parameters of interest.

25. An apparatus for diagnosing faults in a rotating machine, said apparatus comprising:

means for creating a definition of the rotating machine using a plurality of templates that specify components and functional characteristics of the components;

means for identifying a plurality of sensors configured to provide actual operational data related to the rotating machine;

means for defining a plurality of problem rules including logical expressions by which machine faults are deduced from specified portions of the actual operational data; and

means for performing an analysis of operation of the rotating machine, the analysis including acquiring the actual operational data from the plurality of sensors and applying the problem rules and templates to the actual operational data to detect faults in the rotating machine and in response thereto generating an indication when a fault is detected.

26. The apparatus of claim 25, wherein the means for performing an analysis comprises a microprocessor programmed to apply the problem rules and templates to the actual operational data.

27. The apparatus of claim 26, wherein the microprocessor is programmed to acquire data indicating vibration spectra of the rotating machine and apply the problem rules and templates to the actual operational data to search an expected frequency range in the vibration spectra in order to extract a peak amplitude and corresponding fundamental frequency from within the expected frequency range.

28. The apparatus of claim 25, wherein the means for creating a definition of the rotating machine uses at least one of the following templates:

a generic single value for defining a variable of interest in the analysis;

a generic single value equation to define frequency parameters over which variables of interest may be identified;

a base template for modeling characteristics of a component in an unassembled state; and

a component template for modeling characteristics of a component in a power train.

29. A system for diagnosing faults in a rotating machine, the system including:

a microprocessor programmed to perform an analysis of operation of the rotating machine, the analysis including acquiring actual operational data from a plurality of sensors associated with the rotating machine, and applying templates that specify components and functional characteristics of the components to the actual operational data using problem rules to detect faults in the rotating machine and in response thereto generating an indication when a fault is detected.

30. The system of claim 29, wherein the microprocessor is programmed to acquire vibration spectra from the plurality of sensors and apply the problem rules and templates to the actual operational data in order to search an expected frequency range in the vibration spectra and extract a peak amplitude and corresponding fundamental frequency from within the expected frequency range.

31. The system of claim 30, wherein the programmed microprocessor uses a hierarchical tree-based structure to arrange the rules, the tree-based structure allowing an arbitrary grouping of logical expressions.

32. The system of claim 29, wherein the system uses at least one of the following templates:

a generic single value for defining a variable of interest in the analysis;

a generic single value equation to define frequency parameters over which variables of interest may be identified;

a base template for modeling characteristics of a component in an unassembled state; and

a component template for modeling characteristics of a component in a power train.