System and method for combining diagnostic evidences for turbine engine fault detection

• US 20070088982A1
• Filed: 10/18/2006
• Published: 04/19/2007
• Est. Priority Date: 10/18/2005
• Status: Active Grant

• Current Assignee: Honeywell International Inc.
• Original Assignee: Honeywell International Inc.
• Inventors: Mylaraswamy, Dinkar, Guralnik, Valerie, Voges, Harold
• Granted Patent: US 7,337,086 B2
• Time in Patent Office: Days
• US Class Current: 714/26
• CPC Class Codes: G05B 23/0262 Confirmation of fault detection, e.g. extra checks to confirm that a failure has indeed occurred
• CPC Class Codes: G05B 23/0275 Fault isolation and identification, e.g. classify fault; estimate cause or root of failure

Abstract

A system and method for combining conclusions from multiple fault detection techniques to isolate likely faults in a turbine engine is provided. The system and method provide the ability to effectively deal with multiple concurrent faults in the engine. Additionally, the embodiments of the invention provide the ability to correctly characterize multiple conclusions generated from evidence having different levels of interdependence. In one embodiment, the conclusions based on device data with high dependency are aggregated using a high dependency aggregation rule, and the resulting high-dependency sets are then further aggregated using a weak dependency rule. Finally, any conclusions based on independent evidence can be aggregated using an independent combination rule. The resulting aggregation determines which fault(s) are most likely indicated by the plurality of conclusions, taken into account the dependency of the device data used to generate the conclusions.

Claims

1. A fault detection system for detecting faults in a turbine engine, the system comprising:

a plurality of fault detectors, the plurality of fault detectors adapted to receive device data from the turbine engine and determine a plurality of fault conclusions from the device data; and

a diagnostic aggregation mechanism, the diagnostic aggregation mechanism adapted to receive the plurality of fault conclusions and utilize at least one valid multiple fault set to aggregate the plurality of fault conclusions and isolate a likely fault in the turbine engine.

2. The system of claim 1 wherein the diagnostic aggregation mechanism is adapted to aggregate the plurality of fault conclusions and isolate a likely fault in the turbine engine using a hybrid Dezert-Smarandache Theory (DSmT) aggregation rule.

3. The system of claim 1 wherein the at least one valid multiple fault set is determined from a systematic model of the turbine engine, the systematic model describing effects of a plurality of failure modes on the turbine engine.

4. The system of claim 1 wherein the diagnostic aggregation mechanism is further adapted to isolate a likely fault by aggregating the plurality of fault conclusions based on dependency of data used by the plurality of fault detectors to generate the plurality of fault conclusions.

5. The system of claim 4 wherein the diagnostic aggregation mechanism is adapted to aggregate the plurality of fault conclusions by partitioning conclusions into high dependent sets, and further partitioning high dependent sets into weak dependent supersets.

6. The system of claim 5 wherein the diagnostic aggregation mechanism is adapted aggregate conclusions in the high dependent sets using a high dependence fusion rule, and wherein the high dependence fusion rule comprises averaging conclusions based on high dependent device data.

7. The system of claim 6 wherein the diagnostic aggregation mechanism is adapted to aggregate conclusions in the weak dependent super-sets using a weak dependence fusion rule, and wherein the weak dependence fusion rule comprises averaging conclusions using a weighted average.

8. The system of claim 7 wherein the diagnostic aggregation mechanism is further adapted to aggregate conclusions based on independent evidence and the aggregated conclusions in the weak dependent super sets using an independent evidence rule.

9. The system of claim 8 wherein the independent evidence rule comprises a hybrid Dezert-Smarandache Theory (DSmT) aggregation rule.

10. A method of detecting faults in a turbine engine, the method comprising the steps of:

receiving device data from the turbine engine;

determining a plurality of fault conclusions using a plurality of fault detection techniques, the plurality of fault detection techniques each using a subset of the device data; and

isolating likely faults by utilizing at least one valid multiple fault set to aggregate the plurality of fault conclusions and isolate a likely fault in the turbine engine.

11. The method of claim 10 wherein the step of isolating likely faults by utilizing at least one valid multiple fault set to aggregate the plurality of fault conclusions and isolate a likely fault in the turbine engine comprising using a hybrid Dezert-Smarandache Theory (DSmT) aggregation rule.

12. The method of claim 10 wherein the step of isolating likely faults by utilizing at least one valid multiple fault set to aggregate the plurality of fault conclusions and isolate a likely fault in the turbine engine comprises determining the at least one valid multiple fault set using a systematic model of the turbine engine, the systematic model describing effects of a plurality of failure modes on the turbine engine.

13. The method of claim 10 wherein the step of isolating likely faults by utilizing at least one valid multiple fault set to aggregate the plurality of fault conclusions and isolate a likely fault in the turbine engine comprising further isolating likely faults by aggregating the plurality of fault conclusions based on dependency of data used by the plurality of fault detectors to generate the plurality of fault conclusions.

14. The method of claim 13 wherein the step of isolating likely faults comprises partitioning conclusions into high dependent sets, and further partitioning high dependent sets into weak dependent supersets.

15. The method of claim 14 wherein the step of isolating likely faults further comprises the step of aggregating conclusions in the high dependent sets using a high dependence fusion rule, and wherein the high dependence fusion rule comprises averaging conclusions based on high dependent device data.

16. The method of claim 15 wherein the step of isolating likely faults further comprises the step of aggregating conclusions in the weak dependent super-sets using a weak dependence fusion rule, and wherein the weak dependence fusion rule comprises averaging conclusions using a weighted average.

17. The method of claim 16 wherein the step of isolating likely faults further comprises the step of aggregating conclusions based on independent evidence and the aggregated conclusions in the weak dependent super sets using an independent evidence rule.

18. The method of claim 17 wherein the independent evidence rule comprises a hybrid Dezert-Smarandache Theory (DSmT) aggregation rule.

19. A program product comprising:

a) a fault detection program for predicting fault in a turbine engine, the fault detection program including;

a plurality of fault detectors, the plurality of fault detectors adapted to receive device data from the turbine engine and determine a plurality of fault conclusions from the device data; and

a diagnostic aggregation mechanism, the diagnostic aggregation mechanism adapted to receive the plurality of fault conclusions and utilize at least one valid multiple fault set to aggregate the plurality of fault conclusions and isolate a likely fault in the turbine engine; and

b) computer-readable signal bearing media bearing said program.