Self-generating engine-specific health monitoring model from generic model base

Patent metadata

Field	Value
Publication number	US-11300069-B2
Application number	US-201816202142-A
Country	US
Kind code	B2
Filing date	Nov 28, 2018
Priority date	Nov 28, 2018
Publication date	Apr 12, 2022
Grant date	Apr 12, 2022

Abstract

A method for self-generating an engine-specific model in an engine health monitoring system is provided. The method comprises generating a generic engine model including a generic physics-based model for each of a plurality of engine components in the specific engine; capturing a plurality of observed engine component parameters for each of the plurality of engine components and a plurality of observed environmental parameters during one or more pre-planned training missions; and training an engine-specific model using the plurality of observed engine component parameters and the plurality of environmental parameters captured during the one or more pre-planned training missions, wherein the engine-specific model includes an engine-specific physics-based model for each of the plurality of engine components in the specific engine. Each engine-specific physics-based model for an engine component is configured for use in predicting one or more engine component parameters using a second plurality of observed environmental parameters captured during an operational mission.

First claim

What is claimed is:

1. A method for self-generating an engine-specific model in an engine health monitoring system, the method comprising:
identifying a plurality of different types of engine components in an engine in an aircraft and a component layout for the engine;
retrieving a plurality of physics-based component mapping forms from a library of physics-based components that correspond to the identified engine components;
forming a generic engine model based on the identified component layout and the retrieved plurality of physics-based component mapping forms for the plurality of different types of engine components;
capturing, using aircraft sensors, a plurality of observed engine component parameters for each of the plurality of engine components and a plurality of observed environmental parameters for the aircraft operating environment during one or more pre-planned training missions;
applying artificial intelligence and machine learning techniques to train an engine-specific model using the plurality of observed engine component parameters and the plurality of environmental parameters captured during the one or more pre-planned training missions, wherein the engine-specific model is trained through tuning the plurality of physics-based component mapping forms to provide outputs that match the observed engine component parameters under the captured environmental parameter;
wherein each of the plurality of tuned physics-based component mapping forms is configured for use in predicting one or more engine component parameters using a second plurality of observed environmental parameters captured using aircraft sensors during an operational mission performed by the aircraft; and
applying the engine health monitoring system to identify a specific engine component for maintenance action when one or more of a second plurality of observed engine component parameters related to the specific engine component and captured using aircraft sensors during the operational mission deviates from predicted engine component parameters for the specific engine component during the operational mission by a predetermined amount.

2. The method of claim 1 , wherein the identifying the plurality of engine components and the component layout for the engine comprises identifying a number, type, size, and layout of the engine components in the aircraft.

3. The method of claim 2 , wherein the identifying the plurality of engine components and the component layout for the engine comprises determining a number of gear boxes, compressor stages, combustors bleed ports, and turbine stages and an available level of instrumentation.

4. The method of claim 1 , wherein the engine components include a gear box, compressor, combustor, turbine, ducts, nozzle, shaft, secondary flow, and bleeds.

5. The method of claim 1 , wherein the plurality of physics-based component mapping forms comprises a performance map for simulating engine component operation that generates key parameters based on a state of a system of performance maps for the plurality of engine components.

6. The method of claim 5 , wherein the performance maps comprise non-linear, non-empirical component models generated with a theoretical or empirical-based component aero-thermal design tool and without experimental or test data related to the engine.

7. The method of claim 1 , wherein the applying artificial intelligence and machine learning techniques to train an engine-specific model comprises generating sets of scalars for the plurality of physics-based component mapping forms.

8. A system for self-generating an engine specific model in an engine health monitoring system, comprising one or more processors configured by programming instructions encoded on non-transitory computer readable media to:
generate a generic engine model comprising a plurality of physics-based component mapping forms from a library of physics-based components that correspond to a plurality of different types of engine components in a specific engine in an aircraft;
receive a plurality of observed engine component parameters for each of the plurality of engine components and a plurality of observed environmental parameters for the aircraft operating environment captured during one or more pre-planned training missions; and
apply artificial intelligence and machine learning techniques to train an engine-specific model using the plurality of observed engine component parameters and the plurality of environmental parameters captured during the one or more pre-planned training missions, wherein the engine-specific model is trained through tuning the plurality of physics-based component mapping forms to provide outputs that match the observed engine component parameters under the captured environmental parameters;
wherein each of the plurality of tuned physics-based component mapping forms is configured for use in predicting one or more engine component parameters using a second plurality of observed environmental parameters captured during an operational mission performed by the aircraft; and
apply the engine health monitoring system to identify a specific engine component for maintenance action when one or more of a second plurality of observed engine component parameters related to the specific engine component and captured during the operational mission deviates from predicted engine component parameters for the specific engine component during the operational mission by a predetermined amount.

9. The system of claim 8 , wherein the generic engine model generator is further configured to receive information identifying the configuration of the engine which includes information identifying a number, type, size, and layout of the engine components in the vehicle.

10. The system of claim 9 , wherein the information identifying the configuration of the engine comprises a number of compressor stages, combustors, bleed ports, and turbine stages and an available level of instrumentation.

11. The system of claim 8 , wherein the engine components include a gear box, compressor, combustor, turbine, ducts, nozzle, shaft, secondary flow, and bleeds.

12. The system of claim 8 , wherein the plurality of physics-based component mapping forms comprises a performance map for simulating engine component operation that generates key parameters based on a state of a system of performance maps for the plurality of engine components.

13. The system of claim 12 , wherein the performance maps comprise non-linear, non-empirical component models generated with a theoretical or empirical-based component aero-thermal design tool and without experimental or test data related to the engine.

14. The system of claim 8 , wherein to train an engine-specific model the one or more processors are configured to generate sets of scalars for the plurality of physics-based component mapping forms.

15. A non-transitory computer readable media encoded with programming instructions configurable to cause one or more processors in an engine health monitoring system on an aircraft to perform a method, the method comprising:
identifying a plurality of different types of engine components in an engine in an aircraft and a component layout for the engine;
retrieving a plurality of physics-based component mapping forms from a library of physics-based components that correspond to the identified engine components;
forming a generic engine model based on the identified component layout and the retrieved plurality of physics-based component mapping forms for the plurality of different types of engine components;
capturing, using aircraft sensors, a plurality of obs

Assignees

• Honeywell Int Inc

Inventors

• Moeckly Kevin
• Gill Christopher
• Ling Richard

Classifications

• F02D41/22 Primary

  Safety or indicating devices for abnormal conditions {(in air/fuel ratio feedback systems F02D41/1495, in electric control linkage F02D11/107, in purge control systems F02M25/0809)} · CPC title

• G07C5/0808 Primary

  Diagnosing performance data (testing of vehicles G01M17/00; testing of electrical installation on vehicles G01R31/005) · CPC title

• G05B23/0283

  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) · CPC title

• G05B23/0254

  based on a quantitative model, e.g. mathematical relationships between inputs and outputs; functions: observer, Kalman filter, residual calculation, Neural Networks · CPC title

• G06N20/00

  Machine learning · CPC title