Methods and apparatus for the creation and use of reusable fault model components in fault modeling and complex system prognostics

• US 9,959,158 B2
• Filed: 10/13/2015
• Issued: 05/01/2018
• Est. Priority Date: 10/13/2015
• Status: Active Grant

• Current Assignee: Honeywell International Inc.
• Original Assignee: Honeywell International Inc.
• Inventors: Kolandavelu, Raghupathy, Felke, Tim, Kesavan, MoghanaPriya, K, Ramchandra Reddy, Mahoney, Tim
• Primary Examiner(s): Wilson, Yolanda L
• Application Number:
  US14/881,772
• Publication Number:
  US 20170102982A1
• Time in Patent Office: 931 Days
• CPC Class Codes: G05B 17/02 electric
• CPC Class Codes: G05B 23/0248 Causal models, e.g. fault tree; digraphs; qualitative physics
• 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: G06F 11/0751 Error or fault detection not based on redundancy power supply failures G06F1/30; network fault management H04L41/06
• CPC Class Codes: G06F 11/0787 Storage of error reports, e.g. persistent data storage, storage using memory protection
• CPC Class Codes: G06F 11/079 Root cause analysis, i.e. error or fault diagnosis in a hardware test environment G06F11/22; in a software test environment G06F11/36
• CPC Class Codes: G06Q 10/06312 Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
• CPC Class Codes: G06Q 50/04 Manufacturing
• CPC Class Codes: Y02P 90/30 Computing systems specially adapted for manufacturing

Abstract

A method for creating and using a fault model representative of a system, the fault model comprising one or more fault model components is provided. The method creates a reusable fault model component; associates one or more algorithms to the reusable fault model component, the one or more algorithms describing behavior of the reusable fault model component; incorporates the reusable fault model component and the one or more algorithms into the fault model; tunes the reusable fault model component, using available data associated with the fault model; and predicts occurrence of degradation of the system, based on the tuning.

Claims

1. A method for creating and using a fault model representative of a system, the fault model comprising one or more fault model components, the method comprising:

creating a reusable fault model component;

associating one or more algorithms to the reusable fault model component, the one or more algorithms describing behavior of the reusable fault model component, wherein the one or more algorithms comprises at least one of a transfer function, a signal-to-noise ratio, estimated non-linearities, and an error counter;

incorporating the reusable fault model component and the one or more algorithms into the fault model; and

during runtime of the fault model,

recognizing a current regime associated with the reusable fault model component, wherein the current regime provides a context for operation of the reusable fault model component, and wherein the current regime comprises an indication of equipment mode and state affecting a rate at which equipment usage is accumulated, at which fault indications are viewed as legitimate, at which condition indicators are reliable, and at which data needs to be collected;

collecting the data associated with the reusable fault model component, based on the context provided by the current regime;

tuning the reusable fault model component, using available data associated with the fault model; and

predicting occurrence of degradation of the system, based on the collected data and the tuning.

2. The method of claim 1, further comprising:

executing the one or more algorithms to generate performance indices;

detecting at least one trend of performance for the reusable fault model component, wherein the trend is based on the performance indices; and

identifying a time-frame for degradation indicated by the at least one trend;

wherein predicting the occurrence of degradation is based on the time-frame.

3. The method of claim 2, further comprising:

presenting an alert advising maintenance for the system, based on predicting the occurrence of degradation.

4. The method of claim 1, wherein tuning the reusable fault model component further comprises:

initially tuning when the reusable fault model component is integrated into the fault model, the initial tuning using sample data; and

continuously tuning during use of the fault model, the continuously tuning using at least current, real-time data.

5. The method of claim 4, wherein initially tuning the reusable fault model component further comprises:

estimating parameter data associated with the reusable fault model component, based on the sample data;

executing the fault model using the estimated parameter data, to produce updated output; and

determining accuracy of the predicted occurrence of degradation, based on the updated output.

6. The method of claim 5, wherein continuously tuning the reusable fault model component during execution of the fault model further comprises:

estimating parameter data associated with the reusable fault model component, based on the current, real-time data;

executing the fault model using the estimated parameter data, to produce updated output; and

determining the accuracy of the predicted occurrence of degradation, based on the updated output.

7. The method of claim 6, wherein continuously tuning the reusable fault model component during execution of the fault model further comprises:

receiving data recordings for the reusable fault model component;

computing performance indices associated with the reusable fault model component, using input and output data for the reusable fault model component;

estimating the parameter data associated with the reusable fault model component, using the performance indices.

8. The method of claim 6, wherein determining the accuracy of the predicted occurrence of degradation further comprises:

providing the parameter data as input to the fault model;

executing the fault model to generate output; and

updating the one or more algorithms during execution of the fault model, based on the input;

comparing the output with a set of recorded fault data to correlate the parameter data to a set of previous parameter data;

determining accuracy of the predicted occurrence of the degradation, based on the output and the correlated parameter data.

9. The method of claim 1, wherein incorporating the reusable fault model component and the one or more algorithms into the fault model further comprises:

determining a set of required parameters for use of the reusable fault model component in the fault model;

detecting system-defined parameters corresponding to the set of required parameters; and

mapping the set of required parameters to the system-defined parameters.

10. The method of claim 1, wherein incorporating the reusable fault model component and the one or more algorithms into the fault model further comprises:

determining a set of required parameters for use of the reusable fault model component in the fault model, the set of required parameters comprising a subset and a remainder;

detecting a first group of system-defined parameters corresponding to the subset of the set of required parameters;

mapping the subset to the first group of system-defined parameters;

identifying a second group of system-defined parameters corresponding to algorithms applicable to compute at least one value for the remainder; and

computing the at least one value for the remainder, using the algorithms and the second group of system-defined parameters.

11. A method for using reusable sub-assemblies in a fault model associated with a system, the method comprising:

associating one or more algorithms to a reusable sub-assembly comprising a reusable fault model component, the one or more algorithms describing behavior of the reusable fault model component, wherein the one or more algorithms comprises at least one of a transfer function, a signal-to-noise ratio, estimated non-linearities, and an error counter;

incorporating the reusable sub-assembly into the fault model, the reusable sub-assembly representing a component of the system; and

during runtime of the fault model,

recognizing a current regime associated with the reusable fault model component, wherein the current regime provides a context for operation of the reusable fault model component, and wherein the current regime comprises an indication of equipment mode and state affecting a rate at which equipment usage is accumulated, at which fault indications are viewed as legitimate, at which condition indicators are reliable, and at which data needs to be collected;

collecting the data associated with the reusable fault model component, based on the context provided by the current regime;

tuning the reusable fault model component, using available data associated with the fault model; and

predicting potential faults of the system using the reusable sub-assembly, based on the collected data and the tuning, wherein the potential faults indicate degradation of the system.

12. The method of claim 11, further comprising:

estimating parameter data associated with the reusable sub-assembly;

executing the fault model using the estimated parameter data, to produce updated output; and

determining accuracy of the predicted potential faults, based on the updated output.

13. The method of claim 12, further comprising:

receiving data recordings for the reusable sub-assembly, the data recordings obtained at timed intervals;

identifying a transfer function associated with the reusable sub-assembly, using input and output data for the reusable sub-assembly;

estimating the parameter data associated with the reusable sub-assembly, using the data recordings and the transfer function.

14. The method of claim 12, wherein executing the fault model using the estimated parameter data further comprises:

providing the parameter data as input to the fault model;

executing the fault model to generate output; and

during execution of the fault model, updating the one or more algorithms associated with the reusable sub-assembly, based on the input; and

wherein predicting potential faults of the system is based on comparing the output with a set of recorded fault data.

15. The method of claim 11, further comprising:

executing the one or more algorithms to generate performance indices;

detecting at least one trend of performance for the reusable sub-assembly, wherein the trend is based on the performance indices; and

identifying a time-frame for failure of the component indicated by the at least one trend;

wherein predicting potential faults of the system is based on the time-frame, the potential faults comprising failure of the component.

16. The method of claim 15, further comprising:

presenting an alert advising maintenance for the system, based on predicting the potential faults of the system.

17. A non-transitory, computer-readable medium containing instructions thereon, which, when executed by a processor, perform a method comprising:

adjusting parameters associated with a reusable fault model component of a fault model for a system, using real-time data produced by execution of the fault model, by;

associating one or more algorithms to the reusable fault model component, the one or more algorithms describing behavior of the reusable fault model component, wherein the one or more algorithms comprises at least one of a transfer function, a signal-to-noise ratio, estimated non-linearities, and an error counter; and

incorporating the reusable fault model component and the one or more algorithms into the fault model;

executing the fault model using the adjusted parameters, to produce adjusted output; and

during runtime of the fault model,

recognizing a current regime associated with the reusable fault model component, wherein the current regime provides a context for operation of the reusable fault model component, and wherein the current regime comprises an indication of equipment mode and state affecting a rate at which equipment usage is accumulated, at which fault indications are viewed as legitimate, at which condition indicators are reliable, and at which data needs to be collected;

collecting the data associated with the reusable fault model component, based on the context provided by the current regime;

tuning the reusable fault model component, using available data associated with the fault model; and

predicting remaining life of the system, based on the collected data and the tuning.

18. The non-transitory, computer-readable medium of claim 17, wherein the method further comprises:

creating the reusable fault model component, using a reusable component database and user input;

associating one or more algorithms to the reusable fault model component, the one or more algorithms describing behavior of the reusable fault model component, and the parameters comprising data required to execute the one or more algorithms; and

incorporating the reusable fault model component and the one or more algorithms into the fault model.

19. The non-transitory, computer-readable medium of claim 17, wherein incorporating the reusable fault model component into the fault model further comprises:

determining a set of required parameters for use of the reusable fault model component in the fault model, wherein the set of required parameters comprises data required to execute one or more algorithms associated with the reusable fault model component;

detecting system-defined parameters corresponding to the set of required parameters; and

mapping the set of required parameters to the system-defined parameters.

20. The non-transitory, computer-readable medium of claim 17, wherein incorporating the reusable fault model component into the fault model further comprises:

determining a set of required parameters for use of the reusable fault model component in the fault model, wherein the set of required parameters comprises a subset and a remainder, wherein the set of required parameters comprises data required to execute one or more algorithms associated with the reusable fault model component;

detecting a first group of system-defined parameters corresponding to the subset of the set of required parameters;

mapping the subset to the first group of system-defined parameters;

identifying a second group of system-defined parameters corresponding to algorithms applicable to compute at least one value for the remainder; and

computing the at least one value for the remainder, using the algorithms and the second group of system-defined parameters.