METHODS AND SYSTEMS FOR TURBINE LINE REPLACEABLE UNIT FAULT DETECTION AND ISOLATION DURING ENGINE STARTUP

• US 20100303611A1
• Filed: 07/24/2009
• Published: 12/02/2010
• Est. Priority Date: 05/29/2009
• Status: Active Grant

• Current Assignee: Honeywell International Inc.
• Original Assignee: Honeywell International Inc.
• Inventors: Kim, Kyusung
• Granted Patent: US 8,467,949 B2
• Time in Patent Office: Days
• US Class Current: 415/118
• CPC Class Codes: F01D 19/00 Starting of machines or engines; Regulating, controlling, or safety means in connection therewith warming-up before starting F01D25/10; turning or inching gear F01D25/34
• CPC Class Codes: F02C 7/26 Starting; Ignition
• CPC Class Codes: F05D 2230/72 Maintenance
• CPC Class Codes: F05D 2260/80 Diagnostics
• CPC Class Codes: F05D 2260/85 Starting
• CPC Class Codes: G01M 15/14 Testing gas-turbine engines or jet-propulsion engines
• CPC Class Codes: G05B 23/0235 based on a comparison with predetermined threshold or range, e.g. "classical methods", carried out during normal operation; threshold adaptation or choice; when or how to compare with the threshold
• CPC Class Codes: G05B 23/0275 Fault isolation and identification, e.g. classify fault; estimate cause or root of failure

Abstract

Systems and methods for isolating a performance anomaly within one or more line replaceable units (LRUs) on a gas turbine engine by monitoring the start up transient are presented. The system comprises a set of sensors, an anomaly detector and a fault isolation reasoner. Each sensor of the set monitors at least one operating parameter of at least one engine component. The anomaly detector is configured to detect an anomaly in a component by comparing a particular value of an operating parameter to a base line value of that parameter. The specific cause of the startup anomaly is isolated utilizing a set of component reasoners that is based on the nature of the detected anomaly. The key events during the engine startup are identified by the combination of monitoring physically relevant phases of a startup and monitoring the engine control schedule. The values at these key events are used for comparing at the anomaly detector

Claims

1. A method for detecting deterioration in an engine component comprising:

monitoring an engine control unit for the execution of a command associated with an occurrence of a start up event;

recording an engine performance parameter value associated with the start up event at the time the command is executed;

comparing the recorded engine parameter value with a baseline engine parameter value indicating a normal event;

determining if the recorded engine parameter value is materially different than the baseline engine parameter value within a predefined standard; and

if the recorded engine parameter value is materially different that the baseline engine parameter value then generating an anomaly indicator.

2. The method of claim 1, further comprising monitoring the speed (N2) and the measured gas temperature (MGT) of the engine for another event;

and

recording the respective engine performance parameter values of N2 and MGT associated with the other event when the other event occurs.

3. The method of claim 2, wherein the event and the other event comprise at least a partial set of defined events.

4. The method of claim 1, wherein the command is a “

fuel enable”

command.

5. The method of claim 3, wherein the set of defined events comprise an inflection point in a time series of one of N2 and MGT.

6. The method of claim 3, wherein the set of events comprises an inflection point in the derivative time series of one of N2 and MGT.

7. The method of claim 3, wherein the set of events includes engine light-off, Fuel Enable, Peak dMGT/dt, Peak dN2/dt, Peak MGT and Idle.

8. The method of claim 3, wherein the engine performance parameter includes one of time, N2, MGT, dMGT/dt, and dN2/dt.

9. The method of claim 1 wherein the engine component is one of a battery, a fuel system, a starter system and an ignition system.

10. The method of claim 1 wherein the baseline engine parameter value is determined from a baseline time series using principal component modeling analysis.

11. A system for isolating a performance anomaly within one or more engine components comprising:

one or more sensors, each sensor monitoring a different engine parameter time series;

a feature extraction module, whereby a specific subset of event features is extracted from the different engine parameter time series;

a feature selection module, whereby the subset of event features is parsed according to which of the one or more engine components for which it is indicative of performance;

a comparison module, wherein values of the event features are compared to a corresponding baseline value for each event feature contained therein; and

a fault isolation reasoner that identifies a particular anomaly within the one or more engine components based on the comparison.

12. The method of claim 11, wherein the different engine parameter time series are adjusted for ambient temperature.

13. The method of claim 12, wherein the different engine parameter time series are engine speed, measured gas temperature and their time derivatives.

14. The method of claim 13, wherein the subset of events includes a fuel enable event.

15. The method of claim 14, wherein the fuel enable event is determined by monitoring the execution of a set of instructions that when executed control the start up process of the engine.

16. A system for isolating a performance anomaly within one or more engine components comprising:

a set of sensors, each sensor of the set monitoring at least one operating parameter of at least one engine component;

an anomaly detector;

the anomaly detector configured to detect an anomaly in the at least one engine component by comparing a particular value of the at least one operating parameter to a baseline feature value; and

a fault isolation reasoner;

the fault isolation reasoner configured to disambiguate a cause of the anomaly in the at least one engine component utilizing a set of component reasoners that is based on the nature of the detected anomaly.

17. The system of claim 16, wherein the fault isolation reasoner comprises a plurality of subsystem anomaly reasoner modules, each sub-system reasoner module comprising a computer readable medium containing instructions that when executed disambiguate the cause of the anomaly from an ambiguity group specific to an engine subsystem associated with the subsystem anomaly reasoner.

18. The system of claim 17, wherein at least two subsystem anomaly reasoners are logically arranged in a cascading fashion.

19. The system of claim 17, wherein the plurality of subsystem anomaly reasoner modules include a start system anomaly reasoner, an ignition system anomaly reasoner, a fuel system anomaly reasoner and a battery fault isolation reasoner.

20. The system of claim 19, wherein during a start up transient process a fault detected by a start anomaly detector is isolated at a first event occurring during the transient process, a fault detected by an ignition anomaly detector is isolated at a second event occurring during the transient process and a fault detected by a fuel system anomaly detector is isolated at a third event occurring during the transient process.