Method, system, and computer program product for performing prognosis and asset management services

• US 7,328,128 B2
• Filed: 02/22/2006
• Issued: 02/05/2008
• Est. Priority Date: 02/22/2006
• Status: Active Grant

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Goebel, Kai Frank, Bonanni, Pierino Gianni, Moscarino, Gary Paul, Eklund, Neil Holger White
• Primary Examiner(s): Barlow; John
• Assistant Examiner(s): KHUU, HIEN DIEU THI
• Application Number:
  US11/359,339
• Publication Number:
  US 20070198215A1
• Time in Patent Office: 713 Days
• Field of Search: 702/182
• US Class Current: 702/182
• CPC Class Codes: G05B 23/0251 Abstraction hierarchy, e.g. "complex systems", i.e. system is divided in subsystems, subsystems are monitored and results are combined to decide on status of whole system
• CPC Class Codes: G05B 23/0283 Predictive maintenance, e.g. involving the monitoring of a system and, based on the monitoring results, taking decisions on the maintenance schedule of the monitored system; Estimating remaining useful life [RUL] preventive maintenance, i.e. planning maintenance according to the available resources without monitoring the system G06Q10/06

Abstract

A method, system, and computer program product for performing prognosis and asset management services is provided. The method includes calculating an accumulated damage estimate for a component via a diagnostics function and applying future mission data for the component to at least one model that calculates accumulated damage or remaining life of the component. The method also includes inputting the accumulated damage estimate to the model and aggregating damage over time and quality assessments produced by the model. The method further includes calculating a damage propagation profile and remaining life estimate for the component based on the aggregating and providing an uncertainty estimate for the damage estimate and the remaining life estimate.

Claims

1. A method for implementing an integrated prognosis system, comprising:

calculating an accumulated damage estimate for a component via a diagnostics function;

applying future mission information for the component to at least one model that calculates accumulated damage or remaining life, wherein mission information comprises one or more of a power level angle, a mach number and an altitude and wherein the at least one model is tuned using sensor data from a rig test;

inputting the accumulated damage estimate to the at least one model;

assembling data sequences representing only active operation of the rig via a time-mapping function;

removing data resets and negative trending outliers; and

synchronizing specified sensor features to a common, uniformly sampled time vector;

aggregating damage over time and quality assessments produced by the at least one model;

calculating a damage propagation profile and remaining life estimate for the component based upon the aggregating; and

providing an uncertainty estimate for the damage estimate and the remaining life estimate to a user.

2. The method of claim 1, wherein the at least one model includes one of:

a physics-based model; and

an experience-based model.

3. The method of claim 2, wherein the experience-based model is based on using observations from experiments to determine damage propagation for specific operational set points to build a nonlinear general map that allows retrieval of the damage propagation for any operational condition.

4. The method of claim 1, wherein the quality assessments are computed based on at least one of:

a priori performance of the at least one model as a function of operational conditions of the component, the operational conditions comprising at least one of load and speed;

variability of the at least one model during operation;

time elapsed since the at least one model has received a damage estimate update; and

data sources.

5. The method of claim 4, wherein sources of sensor data include debris data, wherein the computation of the quality assessments represents a function of the variability over a previous number of readings, bounded between two limiting values comprising a lower confidence bound and an upper confidence bound when damage is below a specified threshold, and the quality assessments are gated to a fixed value when the damage is above the specified threshold.

6. A method for implementing an integrated prognosis system, comprising:

calculating an accumulated damage estimate for a component via a diagnostics function;

applying future mission information for the component to at least one model that calculates accumulated damage or remaining life, wherein mission information comprises one or more of a power level angle, a mach number and an altitude;

inputting the accumulated damage estimate to the at least one model;

aggregating damage over time and quality assessments produced by the at least one model, such aggregating including aggregating measures of uncertainty, comprising;

calculating at least one probability density function for remaining useful life of the component;

discretizing the at least one probability density function at specified time intervals over a divided universe of discourse;

discounting the discretized probability density function by its time-dependent quality assessment values;

aggregating discounted probability density functions; and

normalizing the aggregated discounted probability density functions at each of the specified time intervals;

calculating a damage propagation profile and remaining life estimate for the component based upon the aggregating; and

providing an uncertainty estimate for the damage estimate and the remaining life estimate to a user.

7. The method of claim 6, wherein a kernel filter is applied to the aggregated discounted probability functions, discounting the aggregated discounted probability density functions in the past more than recent aggregated discounted probability density functions.

8. The method of claim 6, further comprising:

applying an adaptive filter to results of the normalizing operable for reducing non-monotonic behaviors for damage estimate outputs resulting from the at least one probability density function.

9. The method of claim 6, wherein the at least one model includes an experience-based model that is based on using observations from experiments to determine damage propagation for specific operational set points to build a nonlinear general map that allows retrieval of the damage propagation for any operational condition.

10. The method of claim 6, wherein the quality assessments are computed based on at least one of:

a priori performance of the at least one model as a function of operational conditions of the component, the operational conditions comprising at least one of load and speed;

variability of the at least one model during operation;

time elapsed since the at least one model has received a damage estimate update; and

data sources.

11. A computer program product for implementing an integrated prognosis system, the computer program product including a computer readable medium having instructions for executing a method, the method comprising:

calculating an accumulated damage estimate for a component via a diagnostics function;

applying future mission information for the component to at least one model that calculates accumulated damage or remaining life, wherein mission information comprises one or more of a power level angle, a mach number and an altitude and wherein the at least one model is tuned using sensor data from a rig test;

inputting the accumulated damage estimate to the at least one model;

assembling data sequences representing only active operation of the rig via a time-mapping function;

removing data resets and negative trending outliers; and

synchronizing specified sensor features to a common, uniformly sampled time vector;

aggregating damage over time and quality assessments produced by the at least one model;

calculating a damage propagation profile and remaining life estimate for the component based upon the aggregating; and

providing an uncertainty estimate for the damage estimate and the remaining life estimate to a user.

12. The computer program product of claim 11, wherein the at least one model includes one of:

a physics-based model; and

an experience-based model.

13. The computer program product of claim 12, wherein the experience-based model is based on using observations from experiments to determine damage propagation for specific operational set points to build a nonlinear general map that allows retrieval of the damage propagation for any operational condition.

14. The computer program product of claim 11, wherein the quality assessments are computed based on at least one of:

a priori performance of the at least one model as a function of operational conditions of the component, the operational conditions comprising at least one of load and speed;

variability of the at least one model during operation;

time elapsed since the at least one model has received a damage estimate update; and

data sources.

15. The computer program product of claim 14, wherein sources of sensor data include debris data, wherein the computation of the quality assessments represents a function of the variability over a previous number of readings, bounded between two limiting values comprising a lower confidence bound and an upper confidence bound when damage is below a specified threshold, and the quality assessments are gated to a fixed value when the damage is above the specified threshold.

16. A computer program product for implementing an integrated prognosis system, the computer program product including a computer readable medium having instructions for executing a method, the method comprising:

calculating an accumulated damage estimate for a component via a diagnostics function;

applying future mission information for the component to at least one model that calculates accumulated damage or remaining life, wherein mission information comprises one or more of a power level angle, a mach number and an altitude;

inputting the accumulated damage estimate to the at least one model;

aggregating damage over time and quality assessments produced by the at least one model, such aggregating including aggregating measures of uncertainty, comprising;

calculating at least one probability density function for remaining useful life of the component;

discretizing the at least one probability density function at specified time intervals over a divided universe of discourse;

discounting the discretized probability density function by its time-dependent quality assessment values;

aggregating discounted probability density functions; and

normalizing the aggregated discounted probability density functions at each of the specified time intervals;

calculating a damage propagation profile and remaining life estimate for the component based upon the aggregating; and

providing an uncertainty estimate for the damage estimate and the remaining life estimate to a user.

17. The computer program product of claim 16, wherein a kernel filter is applied to the aggregated discounted probability functions, discounting the aggregated discounted probability density functions in the past more than recent aggregated discounted probability density functions.

18. The computer program product of claim 16, further comprising instructions for performing:

applying an adaptive filter to results of the normalizing operable for reducing non-monotonic behaviors for damage estimate outputs resulting from the at least one probability density function.

19. The computer program product of claim 16, wherein the at least one model includes an experience-based model that is based on using observations from experiments to determine damage propagation for specific operational set points to build a nonlinear general map that allows retrieval of the damage propagation for any operational condition.

20. The method of claim 16, wherein the quality assessments are computed based on at least one of:

a priori performance of the at least one model as a function of operational conditions of the component, the operational conditions comprising at least one of load and speed;

variability of the at least one model during operation;

time elapsed since the at least one model has received a damage estimate update; and

data sources.