Method and apparatus for establishing a predictive maintenance database

• US 6,192,325 B1
• Filed: 09/15/1998
• Issued: 02/20/2001
• Est. Priority Date: 09/15/1998
• Status: Expired due to Term

• Current Assignee: Computational Systems Incorporated (Emerson Electric Company)
• Original Assignee: CSI Technology Incorporated (Emerson Electric Company)
• Inventors: Hilemon, Christopher G., Glumac, Miodrag, Rich, Michael D., Reeves, Todd W., Piety, Kenneth R.
• Primary Examiner(s): Hoff, Marc S.
• Assistant Examiner(s): Bui, Bryan
• Application Number:
  US09/153,690
• Time in Patent Office: 889 Days
• Field of Search: 702/182-186, 702/56, 714/5, 714/40, 700/276
• US Class Current: 702/184
• CPC Class Codes: G06N 5/00 Computing arrangements using knowledge-based models

Abstract

The invention provides a computerized method and apparatus which enables a user, even one who has little or no predictive maintenance skills, to establish a predictive maintenance database that defines information needed to monitor equipment in accordance with a predictive maintenance plan. The type of equipment components to be monitored and associated physical characteristics of the components are input to a computer as an equipment configuration, which may include one or more interconnected components. The computer includes a knowledge base that defines relationships between monitoring practices, component types, and physical characteristic information for component types. A predictive maintenance database is constructed for the components using the inference engine operating on the knowledge base, the selected component type, and the selected physical characteristic information. Multiple measurement technologies may be specified for each component. Preferably, for each measurement technology specified, the predictive maintenance database includes measurement points, an analysis parameter set, and an alarm limit set. Equipment configurations may be defined by the user, or they may be stored in a configuration/component warehouse with little or no configuration definition required of the user.

Claims

1. A method for establishing a predictive maintenance database in a computer that defines information needed for a user to monitor components in accordance with a predictive maintenance plan, comprising:

identifying a component type corresponding to a particular component to be monitored;

providing a master file of information that includes at least component identification information and corresponding additional information related to measurements needed by the predictive maintenance plan;

searching the master file for component identification information corresponding to the identified component type to produce at least one set of component identification information and selecting a set of selected component identification information from the at least one set of component identification information; and

constructing database information for a predictive maintenance database for the component to be monitored using the set of selected component identification information and the additional information corresponding to the set of selected component identification information.

2. The method of claim 1 wherein said providing step further comprises:

providing a knowledge base in the computer that defines relationships between monitoring practices, component types and physical characteristic information for component types; and

providing an inference engine in the computer;

and wherein said constructing step further comprises;

operating in part on said knowledge base with said inference engine to construct database information.

3. The method of claim 1 wherein said providing step includes providing component identification information in the form of a model number for components, and wherein said searching step includes selecting a best fit set of component identification information representing the closest relationship between a set of component identification information and the identified component type.

4. The method of claim 1 wherein said searching step includes searching the master file based on model number.

5. The method of claim 1 wherein said searching step includes searching the master file based on physical criteria of the component.

6. A method for establishing a predictive maintenance database in a computer that defines information needed to monitor components in accordance with a predictive maintenance plan, comprising:

identifying a component type corresponding to a particular component to be monitored;

identifying physical characteristic information corresponding to the identified component type and the particular component to be monitored;

providing a knowledge base in the computer that defines relationships between monitoring practices, component types and physical characteristic information for component types;

providing an inference engine in the computer for operating in part on said knowledge base to construct predictive maintenance databases; and

constructing information for a predictive maintenance database for each component to be monitored using the inference engine operating on the knowledge base, the selected component type and the selected physical characteristic information.

7. The method of claim 6 wherein constructing said database information further comprises defining the type of data to be measured for the identified component type and defining measurement points on the component from which data will be measured.

8. The method of claim 6 wherein said constructing step further comprises defining an analysis parameter set including set-up parameters for use by a data collection instrument to collect data in accordance with the predictive maintenance database.

9. The method of claim 6 wherein said constructing step includes defining an alarm limit set including alarm limits delineating normal and abnormal component operation for data measured in accordance with the database.

10. The method of claim 6 wherein said constructing step includes recommending a plurality of measurement points for measuring operating characteristics of the component in accordance with the predictive maintenance plan.

11. The method of claim 10, wherein said constructing step further comprises specifying a recommended type of component operating characteristic to be measured at each of said plurality of measurement points.

12. The method of claim 11, wherein said constructing step further comprises specifying a recommended alarm limit delineating normal and abnormal component operation for a recommended type of component operating characteristic to be measured at a particular one of the plurality of measurement points.

13. The method of claim 10, further comprising:

providing and displaying an image illustrating the component type; and

displaying locations on the component image corresponding to one or more of said plurality of measurement points.

14. The method of claim 6, further comprising obtaining data in accordance with the predictive maintenance database.

15. The method of claim 6, further comprising specifying an operational significance of the selected component type.

16. The method of claim 6, further comprising editing information contained in the knowledge base.

17. The method of claim 6, further comprising storing the identified component type and physical characteristic information as stored user-defined components, and selecting one of the stored user-defined components to thereby identify a component type and physical characteristic information for establishing additional predictive maintenance databases.

18. The method of claim 6 wherein said identifying steps include:

providing a warehouse containing a list of component types and associated physical characteristic information; and

selecting a component type from said list.

19. The method of claim 18, further comprising modifying information contained in the warehouse.

20. The method of claim 6 further comprising specifying a measurement technology for use in monitoring a component type and wherein said constructing step produces a predictive maintenance database corresponding to said measurement technology.

21. A programmable apparatus for establishing a predictive maintenance database defining information needed to monitor components in accordance with a predictive maintenance plan, the apparatus comprising:

a memory having a knowledge base defining relationships between monitoring practices, component types and physical characteristic information for component types;

a data processor having an inference engine for operating in part on said knowledge base to construct database information for a predictive maintenance database; and

a user interface for inputting user commands to the data processor including commands that;

(a) identify a component type corresponding to a particular component to be monitored;

(b) identify physical characteristic information corresponding to the identified component type and the particular component to be monitored; and

said data processor further comprising means for constructing information for a predictive maintenance database for each component to be monitored using the inference engine operating on the knowledge base, the selected component type and the selected physical characteristic information.

22. The apparatus of claim 21 wherein said data processor is further operable to operate in part on said knowledge base to construct a predictive maintenance database defining measurement points, monitoring schedules, data to be measured, and alarm limits delineating normal and abnormal component operation to be applied to the measured data.

23. The apparatus of claim 21 wherein said data processor is further operable to construct a predictive maintenance database by defining an analysis parameter set which includes set-up parameters for use by a data collection instrument to collect data in accordance with the predictive maintenance plan.

24. The apparatus of claim 21 wherein said data processor is further operable to construct a predictive maintenance database by defining an alarm limit set including a plurality of alarm limits delineating normal and abnormal component operation for data measured in accordance with the database.

25. The apparatus of claim 21, further comprising:

a display connected to said data processor for displaying the images of the component types.

26. A method for establishing a predictive maintenance database in a computer that defines information needed for a user to monitor components in accordance with a predictive maintenance plan, comprising:

displaying a plurality of component groups, prompting a user to select at least one of the component groups, and receiving a user input identifying a selected component group corresponding to a particular component to be monitored;

displaying a plurality of component types corresponding to the selected group, prompting the user to select at least one of the component types, and receiving a user input identifying a selected component type corresponding to the particular component to be monitored;

displaying physical characteristic types corresponding to the selected component type, prompting the user to provide physical characteristic information corresponding to the physical characteristic types to further define the particular component to be monitored, and accepting physical characteristic information provided by the user;

providing a knowledge base in the computer that defines relationships between monitoring practices and component types and physical characteristic information for component types;

providing an inference engine in the computer for constructing database information for a predictive maintenance database based in part on said knowledge base;

using the knowledge base to select measurement specifications based on the selected component type and the physical characteristic information provided by the user; and

constructing database information for the predictive maintenance database for the particular component to be monitored using the inference engine operating on the knowledge base, the selected component type, the selected physical characteristic information, and the user defined measurement specifications.

27. The method of claim 26, further comprising prompting a user to accept or modify the measurement specifications and producing user defined measurement specifications based on the user'"'"'s response to said prompting.

28. The method of claim 26, further comprising obtaining predictive maintenance data in accordance with the predictive maintenance database constructed for the components to be monitored.

29. The method of claim 28 further comprising the step of comparing predictive maintenance data to a criterion, and indicating an alarm condition when the criterion is met.

30. The method of claim 26 further comprising:

defining the type of data to be measured for the identified component type, and constructing the predictive maintenance database to correspond to the type of data.

31. The method of claim 26 wherein said constructing step includes defining an analysis parameter set including set-up parameters for use by a data collection instrument to collect data in accordance with the predictive maintenance database.

32. The method of claim 26 wherein said constructing step includes defining an alarm limit set including alarm limits delineating normal and abnormal component operation for data measured in accordance with the database.

33. The method of claim 26, further comprising storing the selected component type and the selected physical characteristic information for use in establishing further predictive maintenance databases.

34. A method for establishing a predictive maintenance database in a computer that defines information needed for a user to monitor a configuration of components defined by a plurality of interconnected components in accordance with a predictive maintenance plan, comprising:

defining a configuration of components by;

displaying a first plurality of component groups,

prompting a user to select at least one of the component groups to produce a first selected group,

receiving a user input identifying a first selected component group, corresponding to a first component in the configuration;

displaying a first plurality of component types corresponding to the first selected group,

prompting the user to select at least one of the component types to produce a first selected component type,

receiving a user input identifying a first selected component type corresponding to the first component;

displaying physical characteristic types corresponding to the first selected component type,

prompting the user to provide physical characteristic information corresponding to the physical characteristic types to further define the first selected component type,

accepting physical characteristic information provided by the user;

displaying a second plurality of component groups,

prompting a user to select at least one of the component groups to produce a second selected group,

receiving a user input identifying a second selected component group corresponding to a second component in the configuration;

displaying a second plurality of component types corresponding to the second selected group,

prompting the user to select at least one of the component types to produce a second selected component type,

receiving a user input identifying a second selected component type corresponding to the second component;

displaying physical characteristic types corresponding to the second selected component type,

prompting the user to provide physical characteristic information corresponding to the physical characteristic types to further define the second selected component type,

accepting physical characteristic information provided by the user corresponding to the second selected component type;

displaying a plurality of spatial orientations, interconnection arrangements, and types of couplings,

prompting the user to select a spatial orientation, interconnection arrangement, and coupling type,

receiving user inputs identifying a selected spatial orientation, interconnection arrangement, and coupling type to define a selected configuration corresponding to a physical interconnection between the first and second selected component types;

providing a knowledge base in the computer that defines relationships between monitoring practices, component types, and physical characteristic information for configuration types;

providing an inference engine in the computer for constructing database information for a predictive maintenance database based in part on said knowledge base;

using the knowledge base to select measurement specifications based on the selected configuration and the physical characteristic information provided by the user; and

constructing database information for the predictive maintenance database for the particular configuration to be monitored using the inference engine operating on the knowledge base, the selected configuration, the selected physical characteristic information, and the user defined measurement specifications.

35. The method of claim 34, further comprising storing the selected configuration, the selected physical characteristic information, and the user defined measurement specifications for establishing further predictive maintenance databases.

36. A data processing apparatus for defining component configurations, the data processing apparatus including a user interface for receiving commands and data from a user and comprising:

a component design studio for displaying a plurality of component types, for responding to user inputs, to select a component type and for displaying the selected component type;

a first user interface screen responsive to user input and the selected component type for prompting the user to provide physical parameter information related to the selected component type and for accepting parameter information provided by the user; and

processing means for defining a component configuration based upon at least the selected component type and the parameter information.

37. The apparatus of claim 36 further comprising:

a second user interface screen responsive to the selected component type for prompting the user to provide analysis information related to analysis of data and for accepting the analysis information that is input by the user; and

wherein said processing means is responsive to the selected component type, the parameter information and the analysis information for defining a component configuration.

38. The apparatus of claim 36 wherein said component design studio further comprises means for displaying a plurality of predictive measurement technologies and for responding to user inputs to select at least one predictive maintenance technology.

39. The apparatus of claim 38 when said first user interface screen is responsive to the selected predictive maintenance technology for prompting the user to provide parameter information corresponding to the selected predictive maintenance technology.

40. The apparatus of claim 36 further comprising:

a second user interface screen responsive to user input and the selected component type for prompting the user to provide analysis information related to analysis of data and for accepting analysis information;

said processing means being responsive to the selected component type, the parameter information and the analysis information for defining a component configuration;

said component design studio further comprising means for displaying a plurality of predictive maintenance technologies and for responding to user inputs to select at least one predictive maintenance technology; and

said first user interface screen and said second user interface screen being responsive to the selected predictive maintenance technology for prompting the user to provide, respectively, parameter information and analysis information corresponding to the selected predictive maintenance technology.

41. The apparatus of claim 36 wherein said computer design studio graphically displays a plurality of component types including couplings for being selected by a user, selects a plurality of component types based on user input, graphically displays the plurality of selected component types, graphically represents the position of each selected component type to each of the other selected component types and graphically represents couplings connected between other component types.

42. A method for graphically associating a plurality of machine components in a computer to define an equipment configuration for use in establishing a predictive maintenance database for the equipment configuration, comprising:

identifying a first component type corresponding to a first component to be monitored, said first component type having a plurality of first physical component parameters;

specifying first component information corresponding to one or more of said plurality of first physical component parameters;

producing a first component configuration from the identified first component type and first component information;

identifying a second component type corresponding to a second component to be monitored, said second component type having a plurality of second physical component parameters;

specifying second component information corresponding to one or more of said plurality of second physical component parameters;

producing a second component configuration from the identified second component type and second component information;

defining a physical coupling between the first component type and the second component type; and

producing an equipment configuration from the first component configuration, the second component configuration, and the physical interconnection.

43. The method of claim 42, further comprising:

associating with the first component type at least one of a plurality of measurement technologies to produce one or more selected measurement technologies; and

constructing a predictive maintenance database based on the identified first component type, the first component information, and said one or more selected measurement technologies.

44. A method for establishing a predictive maintenance database in a computer that defines information needed to monitor components in accordance with a predictive maintenance plan, comprising:

identifying a component type corresponding to a particular component to be monitored;

identifying physical characteristic information corresponding to the identified component type and the particular component to be monitored;

providing a set of rules in the computer which define relationships between monitoring practices, component types and physical characteristic information for component types;

associating a type of data to be collected with the identified component type and the identified physical characteristic information; and

constructing a predictive maintenance database for the particular component to be monitored using the set of rules operating on the identified component type, the identified physical characteristic information, and the type of data to be collected.

45. The method of claim 44 wherein said step of associating includes associating a vibration measurement technology as the type of data to be collected.

46. A programmable apparatus for establishing a predictive maintenance database defining information needed to monitor components in accordance with a predictive maintenance plan, the apparatus comprising:

a memory having a master file of information in the computer that includes at least component identification information and corresponding predictive maintenance database information that specifies the types of measurements needed by the predictive maintenance plan for each component in the master file;

a user interface for inputting user commands including commands that provide component identification information and thereby identify components to be monitored;

a data processor for receiving commands from the user interface and having a search engine for searching the master file for component identification information corresponding to an identified component to produce at least one set of component identification information and selecting a set of selected component identification information from the at least one set of component identification information; and

said data processor further comprising means for constructing database information for a predictive maintenance database for an identified component to be monitored using the set of selected component identification information and the predictive database information corresponding to the set of selected component identification information.

47. The apparatus of claim 46 wherein said component identification information is selected from the following group:

a manufacturer name for components, a model number for component, and physical criteria corresponding to physical characteristics of components.

48. The apparatus of claim 46 wherein said predictive database information is selected from the following group:

a type of measurement analysis to be performed in accordance with the predictive maintenance plan, measurement point locations identifying points on components where predictive maintenenance data is to be measured in accordance with the predictive maintenance plan, and spectral analysis parameters to be employed for collection of spectral data.

49. The apparatus of claim 46 wherein said search engine further comprises means for automatically selecting a best fit set of component identification information representing the closest relationship between a set of component identification information and the identified component type.

50. The apparatus of claim 46 wherein said search engine further comprises means for producing a plurality of records of component identification information and selecting a best fit set of component identification information from the plurality of records of component identification information.

51. The apparatus of claim 46 wherein said search engine further comprises means for searching the master file based on at least one of the following group:

manufacturer name, model number, and physical criteria of the component.