Dynamically configurable process for diagnosing faults in rotating machines

• US 20020138217A1
• Filed: 03/20/2001
• Published: 09/26/2002
• Est. Priority Date: 03/20/2001
• Status: Active Grant

• 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.
• Granted Patent: US 6,560,552 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 15/12 by monitoring vibrations
• CPC Class Codes: G01R 31/343 in operation

Abstract

A technique for detecting problems in a rotating machine is configured by the user specifying components and the functional characteristics of a particular installation of a rotating machine. The machine specification are used to identify salient observations that can be derived from measurement data regarding the machine performance. Rule are specified by the user to deduce machine signatures from the salient observations and detect problems evidenced by the machine signatures.

Claims

1. A method by which an apparatus diagnoses faults in a rotating machine, said method comprising:

storing, in the apparatus, a definition of the rotating machine which specifies components and functional characteristics of the components;

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

storing specification of portions of the operational data which are to be selected for analysis;

storing rules that define formulae by which machine faults are deduced from the portions of the operational data; and

performing an analysis of operation of the rotating machine by;

(a) acquiring actual operational data from the plurality of sensors,

(b) employing the specification of portions of the operational data to extract portions of the actual operational data,

(c) applying the rules to extracted portions of the actual operational data to detect faults in the rotating machine and in response thereto generate an indication when a fault is detected.

2. The method as recited in claim 1 wherein storing a definition of the rotating machine comprises:

storing, in the apparatus, a table of components of the rotating machine; and

modifying the table of components with data defining operational characteristics of the components to produce a manufacturer specific definition of each component.

3. The method as recited in claim 2 wherein storing a definition of the rotating machine further comprises incorporating site specific features with the manufacturer specific definition of each component to produce a site specific definition of the rotating machine.

4. The method as recited in claim 1 wherein storing a definition of the rotating machine comprises incorporating site specific features to produce a site specific definition of the rotating machine.

5. The method as recited in claim 4 wherein the rules are specified utilizing the site specific definition of the rotating machine.

6. The method as recited in claim 1 wherein storing rules comprises:

storing signature rules that define formulae by which machine performance characteristics are deduced from the portions of the operational data; and

storing problem rules that define formulae by which machine faults are deduced from the machine signatures.

7. The method as recited in claim 6 wherein the signature rules are specified utilizing the definition of the rotating machine.

8. The method as recited in claim 6 wherein applying the rules comprises:

applying the machine signatures to extracted portions of the actual operational data to produce salient observations regarding characteristics of the rotating machine; and

applying the problem rules to the salient observations, and in response thereto, generating an indication of a machine fault.

9. The method recited in claim 6 wherein acquiring actual operational data comprises acquiring data indicating vibration of the rotating machine.

10. A method by which an apparatus diagnoses faults in a rotating machine to which the apparatus is connected, said method comprising:

storing, in the apparatus, a table of components of the rotating machine;

modifying the table of components with data defining operational characteristics of the components to produce a manufacturer specific definition of each component;

incorporating site specific features with the manufacturer specific definition of each component to produce a site specific definition of the rotating machine;

identifying a plurality of sensors attached to the rotating machine which provide operational data related to the rotating machine;

storing, in the apparatus, a specification of portions of the operational data which are to be selected for analysis;

storing machine signature rules that define performance characteristics of the machine which can be deduced from the portions of the operational data, wherein the machine signatures are specified using the site specific definition of the rotating machine;

storing, in the apparatus, problem rules that define faults which can be deduced from the machine signatures, wherein the problem rules are specified using the site specific definition of the rotating machine; and

performing an analysis of operation of the rotating machine by;

(a) acquiring actual operational data from the plurality of sensors,

(b) employing the specification of portions of the operational data to extract portions of the actual operational data;

(c) applying the machine signatures to extracted portions of the actual operational data to produce salient observations about the rotating machine; and

(d) applying the problem rules to the salient observations and in response thereto generating an indication of a machine fault.

11. A method by which an apparatus diagnoses faults in a rotating machine to which the apparatus is connected, said method comprising:

storing, in the apparatus, a definition of a plurality of components of the rotating machine which definition specifies operational characteristics of each component;

incorporating site specific features with the definition of a plurality of components to produce a site specific definition of the rotating machine;

providing an identification of a plurality of vibration sensors attached to the machine and to the apparatus, wherein each of the plurality of vibration sensors produces a vibration signal;

storing, in the apparatus, a specification of a portion each vibration signal that is to be selected for analysis;

wherein specification of specification of a portion each vibration signal is based on data from the site specific definition of the rotating machine;

storing machine signature rules that define performance characteristics of the machine that can be deduced from the portions of the operational data, wherein specification of machine signatures utilizes the site specific definition of the rotating machine;

storing, in the apparatus, problem rules that define machine faults that can be deduced from the machine signatures, wherein the specification of problem rules utilizes the site specific definition of the rotating machine; and

performing an analysis of operation of the rotating machine by;

(a) acquiring actual operational data from the plurality of sensors,

(b) employing the specification of portions of the operational data to extract portions of the actual operational data;

(c) applying the machine signatures to extracted portions of the actual operational data to produce salient observations about the rotating machine; and

(d) applying the problem rules to the salient observations and in response thereto generating an indication of a machine fault.

12. The method as recited in claim 11 wherein storing a definition of a plurality of components comprises:

storing, in the apparatus, a table of components of the rotating machine; and

modifying the table of components with data defining operational characteristics of the components to produce a manufacturer specific definition of each component.

13. The method as recited in claim 11 wherein the step of acquiring comprises converting the actual operational data into frequency domain data.

14. The method as recited in claim 11 wherein the step of acquiring converts the actual operational data into frequency domain data, and the specification of a portion each vibration signal specifies a range of frequencies to be analyzed in the frequency domain data.