Predictive aircraft maintenance systems and methods incorporating classifier ensembles

• US 10,239,640 B2
• Filed: 12/11/2015
• Issued: 03/26/2019
• Est. Priority Date: 12/11/2015
• Status: Active Grant

• Current Assignee: The Boeing Co.
• Original Assignee: The Boeing Co.
• Inventors: Ethington, James M., Sturlaugson, Liessman E., Schimert, James, Wilmering, Timothy J.
• Primary Examiner(s): Kleinman, Lail A
• Application Number:
  US14/966,372
• Publication Number:
  US 20170166328A1
• Time in Patent Office: 1,201 Days
• Field of Search: None
• CPC Class Codes: B64F 5/60 Testing or inspecting aircraft components or systems
• CPC Class Codes: G06Q 10/06311 Scheduling, planning or task assignment for a person or group
• CPC Class Codes: G06Q 10/20 Administration of product repair or maintenance
• CPC Class Codes: G07C 5/0841 Registering performance data recording measured values G01D; information storage G11B

Abstract

Predictive aircraft maintenance systems and methods are disclosed. Predictive maintenance methods may include extracting feature data from flight data collected during a flight of the aircraft, applying an ensemble of related classifiers to produce a classifier indicator for each classifier of the ensemble of classifiers, aggregating the classifier indicators to produce an aggregate indicator indicating an aggregate category of a selected component for a threshold number of future flights, and determining the performance status of the selected component based on the aggregate indicator. The classifiers are each configured to indicate a category of the selected component within a given number of flights. The given number of flights for each classifier is different. The threshold number of future flights is greater than or equal to the maximum of the given numbers of the classifiers. Predictive maintenance systems may include modules configured to extract feature data, classify feature data, and aggregate classifications.

Claims

1. A method (100), executed on a predictive aircraft maintenance computer system (10), of determining a performance status of a selected component (24) in an aircraft (20), the method comprising:

extracting (104), by the predictive aircraft maintenance computer system, feature data from flight data collected during a flight of the aircraft, wherein the feature data relates to performance of one or more components (24) of the aircraft;

applying (106), by the predictive aircraft maintenance computer system, an ensemble of related classifiers (66) configured to identify categories to which the feature data belong to produce a classifier indicator for each classifier of the ensemble of related classifiers, wherein each classifier is configured to indicate a category of the selected component of the aircraft within a given number of future flights, and wherein the given number for each classifier is different;

aggregating (108), by the predictive aircraft maintenance computer system, the classifier indicators to produce an aggregate indicator that indicates an aggregate category of the selected component for a threshold number of future flights, wherein the threshold number is greater than or equal to a maximum of the given numbers of the classifiers;

determining (110), by the predictive aircraft maintenance computer system, the performance status of the selected component relative to the threshold number of future flights based on the aggregate indicator; and

based on the performance status indicating a predicted impending non-performance event of the selected component, causing a maintenance schedule for the aircraft to be modified, wherein modifying the maintenance schedule comprises generating a schedule event to have the selected component repaired prior to the predicted impending non-performance event.

2. The method of claim 1, wherein the component is at least one of an actuator, a valve (28), an air regulator, a primary bleed air pressure regulator, and a shutoff valve.

3. The method of claim 1, wherein the extracting includes determining a statistic of flight data during a time window.

4. The method of claim 1, wherein the extracting includes determining a difference of sensor values during a time window and wherein the flight data includes the sensor values.

5. The method of claim 1, wherein the extracting includes determining a difference between a first sensor value and a second sensor value, and wherein the flight data includes the first sensor value and the second sensor value.

6. The method of claim 1, wherein each classifier indicator indicates either the impending non-performance event of the selected component or no impending non-performance event of the selected component.

7. The method of claim 1, wherein the given numbers of the classifiers form a sequence of consecutive integers beginning with 1.

8. The method of claim 1, wherein the aggregating includes setting the aggregate indicator to one of a maximum value of the classifier indicators, a most common value of the classifier indicators, and a cumulative value of the classifier indicators.

9. The method of claim 1, wherein the aggregating includes classifying each classifier indicator as one of two states, wherein the states include an impending-non-performance state and a likely-performance state, and wherein the aggregating includes setting the aggregate indicator to a most common state of the classifier indicators.

10. A method (100), executed on a predictive aircraft maintenance computer system, of preventive maintenance for an aircraft (20), the method including:

performing the method of claim 1 with the predictive aircraft maintenance computer system; and

determining, with the predictive aircraft maintenance computer system, whether to repair the selected component before the threshold number of future flights.

11. A predictive aircraft maintenance computer system (10) comprising:

at least one processing unit (202);

a computer-readable memory (206) storing non-transitory computer-readable instructions (220) that, when executed by the at least one processing unit, cause the predictive aircraft maintenance computer system to;

extract feature data from flight data collected during a flight of an aircraft (20), wherein the feature data relates to performance of one or more components of the aircraft;

produce a classifier indicator for each classifier of an ensemble of related classifiers (66), wherein each classifier is configured to indicate a category of a selected component (24) of the aircraft within a given number of future flights based upon the feature data, and wherein the given number for each classifier is different;

produce an aggregate indicator that indicates a performance category of the selected component for a threshold number of future flights, wherein the threshold number is greater than or equal to a maximum of the given numbers of the classifiers; and

modify, based on the aggregate indicator indicating a predicted impending non-performance event of the selected component, a maintenance schedule for the aircraft to have the selected component repaired prior to the predicted impending non-performance event.

12. The predictive aircraft maintenance computer system of claim 11, further comprising a data link (74) configured to communicate with a flight data storage system (50), and wherein the flight data storage system is on board the aircraft.

13. The predictive aircraft maintenance computer system of claim 11, further comprising a display, wherein the display is configured to indicate the aggregate indicator with at least one of a visual display, an audio display, and a tactile display.

14. The predictive aircraft maintenance computer system of claim 11, wherein the selected component is at least one of an actuator, a valve, an air regulator, a primary bleed air pressure regulator, and a shutoff valve.

15. The predictive aircraft maintenance computer system of claim 11, wherein the instructions, when executed by the at least one processing unit, further cause the predictive aircraft maintenance computer system to determine a statistic of flight data during a time window.

16. The predictive aircraft maintenance computer of claim 11, wherein the given numbers of the classifiers form a sequence of consecutive integers beginning with 1.

17. The predictive aircraft maintenance computer system of claim 11, wherein the ensemble of related classifiers includes a first classifier (66) with a given number of 1 and a second classifier (66) with a given number of 2.

18. The predictive aircraft maintenance computer system of claim 11, wherein each classifier is the result of guided machine learning.

19. The predictive aircraft maintenance computer system of claim 11, wherein the instructions, when executed by the at least one processing unit, further cause the predictive aircraft maintenance computer system to set the aggregate indicator to one of a maximum value of the classifier indicators, a most common value of the classifier indicators, and a cumulative value of the classifier indicators.

20. The predictive aircraft maintenance computer system of claim 11, wherein the instructions, when executed by the at least one processing unit, further cause the predictive aircraft maintenance computer system to:

classify each classifier indicator as one of two states, wherein the states include an impending-non-performance state and a likely-performance state; and

set the aggregate indicator to a most common state of the classifier indicators.