Failure detection system risk reduction assessment

• US 8,214,317 B2
• Filed: 08/17/2009
• Issued: 07/03/2012
• Est. Priority Date: 08/17/2009
• Status: Active Grant

• Current Assignee: Aerojet Rocketdyne Of De, Inc. (L3Harris Technologies, Inc.)
• Original Assignee: Pratt & Whitney Rocketdyne Incorporated (L3Harris Technologies, Inc.)
• Inventors: Aguilar, Robert B., Huang, Zhaofeng
• Primary Examiner(s): Chen, Alan
• Application Number:
  US12/542,241
• Publication Number:
  US 20110040719A1
• Time in Patent Office: 1,051 Days
• Field of Search: None
• US Class Current: 706/52
• CPC Class Codes: F01P 11/18 concerning coolant pressure, coolant flow, or liquid-coolant level
• CPC Class Codes: F01P 2031/18 Detecting fluid leaks
• CPC Class Codes: G05B 23/024 Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks
• CPC Class Codes: G06F 11/008 Reliability or availability analysis

Abstract

A process includes determining a probability of a failure mode of a system being analyzed reaching a failure limit as a function of time to failure limit, determining a probability of a mitigation of the failure mode as a function of a time to failure limit, and quantifying a risk reduction based on the probability of the failure mode reaching the failure limit and the probability of the mitigation.

Claims

1. A process comprising:

determining a probability of a failure mode of a system being analyzed reaching a failure limit as a function of time to failure limit;

determining a probability of a mitigation of the failure mode as a function of a time to failure limit; and

quantifying a risk reduction based on the probability of the failure mode of reaching the failure limit and the probability of the mitigation.

2. A process comprising:

determining a probability of a failure mode of a system being analyzed reaching a failure limit as a function of time to failure limit;

determining a probability of a mitigation of the failure mode as a function of a time to failure limit; and

quantifying a risk reduction based on the probability of the failure mode of reaching the failure limit and the probability of the mitigation,

wherein said step of determining a probability of a failure of a system being analyzed as a function of time to failure limit includes;

a) determining a risk probability for at least one failure mode, the risk probability corresponding to a likelihood that the at least one failure mode will occur in the system being analyzed;

b) determining a correlation ranking for at least one failure signature, the correlation ranking corresponding to a likelihood that the at least one failure signature represents the at least one failure mode;

c) determining a failure mode risk probability for the at least one failure signature comparing the correlation ranking of the at least one failure signature to a sum of all correlation rankings for a signature set, the signature set including at least the at least one failure signature; and

d) determining a distribution corresponding to a probability that the at least one failure signature will occur at a given time to failure limit in the system being analyzed.

3. The method of claim 2, wherein said step (a) is performed in response to historical data of actual failures, predicted data of future failures, or both.

4. The method of claim 2, wherein said step (a) includes:

predicting a cluster of systems being analyzed;

determining a probability that the failure mode will occur in each system in the cluster; and

averaging the probabilities to determine an overall probability that the failure mode will occur in the system being analyzed.

5. The system of claim 2, wherein the correlation ranking of said step (b) is in a range from 0-5.

6. The method of claim 2, wherein said step (d) is only performed for a selected failure signature if the selected failure signature has a non-zero correlation ranking.

7. The method of claim 2, wherein the system being analyzed is a machine or a process.

8. The method of claim 2, wherein said step of determining a probability of a mitigation of the failure as a function of a time to failure limit includes:

e) determining an effectiveness score corresponding to an ability to respond to the at least one failure signature using a failure detection system to prevent the at least one failure mode from reaching a failure limit in the system being analyzed as a function of time to failure limit.

9. The method of claim 8, wherein said step of quantifying a risk reduction based on the probability of the failure and the probability of the mitigation includes:

f) calculating a failure limit occurrence probability reductions for the at least one signature in response to the effectiveness score;

g) calculating a sum of failure limit occurrence probability reductions from said step (f) for each signature in the signature set; and

h) comparing the sum of failure limit occurrence probability reductions for the failure detection system to a sum of overall failure limit occurrence probabilities in the system being analyzed without implementation of the failure detection system to determine a risk reduction fraction of the failure detection system.

10. The method of claim 9, wherein said step (f) includes integrating on the computer a product of a function corresponding to the distribution of said step (d) and a function corresponding to the effectiveness score of said step (e).

11. The method of claim 9, wherein the failure detection system is a first failure detection system, and wherein said step (h) includes dividing the sum of failure limit occurrence probabilities for the first failure detection system by a sum of failure limit occurrence probabilities for a second failure detection system.

12. The method of claim 8, wherein said step of determining a probability of a mitigation of the failure as a function of a time to failure limit includes:

f) calculating a failure limit occurrence probability reduction for the at least one signature in response to the effectiveness score;

g) selectively repeating steps (a)-(f) for a plurality of failure modes;

h) calculating a sum of failure limit occurrence probability reductions from said step (g) for each signature in the signature set; and

i) comparing the sum of failure limit occurrence probability reductions for the failure detection system to a sum of overall failure limit occurrence probabilities in the system being analyzed without implementation of the failure detection system to determine a risk reduction of the failure detection system.

13. A computer-implemented system, comprising:

a storage module storing at least one failure mode for a system being analyzed; and

a microprocessor, the microprocessor being operable to determine a probability of the at least one failure mode of the system being analyzed reaching a failure limit as a function of time to failure limit, determine a probability of a mitigation of the failure mode as a function of a time to failure limit, and quantify a risk reduction based on the probability of the failure mode reaching the failure limit and the probability of the mitigation.

14. The system of claim 13, wherein an input/output module is also operable to receive a plurality of failure signatures, and to output the risk reduction.

15. A computer-implemented system, comprising:

a storage module storing at least one failure mode for a system being analyzed;

a microprocessor, the microprocessor being operable to determine a probability of the at least one failure mode of the system being analyzed reaching a failure limit as a function of time to failure limit, determine a probability of a mitigation of the failure mode as a function of a time to failure limit, and quantify a risk reduction based on the probability of the failure mode reaching the failure limit and the probability of the mitigation; and

an input/output module operable to receive a plurality of failure signatures, and to output the risk reduction,

wherein the microprocessor determines a probability of a failure of a system being analyzed as a function of time to failure limit by determining a risk probability for at least one failure mode, determining a correlation ranking for at least one failure signature, determining a failure mode risk probability for the at least one failure signature comparing the correlation ranking of the at least one failure signature to a sum of all correlation rankings for a signature set, and determining a distribution corresponding to a probability that the at least one failure signature will occur at a given time to failure limit in the system being analyzed.

16. The system of claim 15, wherein the microprocessor determines a probability of a mitigation of the failure as a function of a time to failure limit by determining an effectiveness score corresponding to an ability to respond to the at least one failure signature using a failure detection system to prevent the at least one failure mode from reaching a failure limit in the system being analyzed as a function of time to failure limit.

17. The system of claim 16, wherein the microprocessor quantifies a risk reduction based on the probability of the failure and the probability of the mitigation by calculating a failure limit occurrence probability reduction for the at least one signature in response to the effectiveness score, calculating a sum of failure limit occurrence probability reductions for each signature in the signature set, and comparing the sum of failure limit occurrence probability reductions for the failure detection system to a sum of overall failure limit occurrence probabilities in the system being analyzed without implementation of the failure detection system to determine a risk reduction of the failure detection system.