Prediction methods and systems for structural repair during heavy maintenance of aircraft

Patent metadata

Field	Value
Publication number	US-2017372000-A1
Application number	US-201615192700-A
Country	US
Kind code	A1
Filing date	Jun 24, 2016
Priority date	Jun 24, 2016
Publication date	Dec 28, 2017
Grant date	—

Abstract

Hybrid feature selection methods include methods of creating a predictive model for structural repair during heavy maintenance in a fleet of aircraft. Methods include qualifying a qualification dataset of fatigue-related parameters calculated from data collected during a first group of flights of a first aircraft that experience a replacement of a structural component during heavy maintenance. Methods include receiving a qualified selection of the fatigue-related parameters and verifying a verification dataset of the qualified selection of the fatigue-related parameters calculated from data collected during a second group of flights of a second aircraft that experienced heavy maintenance without replacement of the structural component. Methods include receiving a set of verified and qualified fatigue-related parameters and building a predictive model for structural repair during heavy maintenance with a training dataset of the verified and qualified fatigue-related parameters calculated from data collected during additional flights of the fleet.

First claim

1 . A method for creating a predictive model for structural repair during heavy maintenance in a fleet of aircraft that each includes a structural component, the method comprising:
for a first aircraft of the fleet that experienced a replacement of the structural component of the first aircraft during heavy maintenance of the first aircraft:
calculating a qualification dataset of fatigue-related parameters from data collected during a first group of flights of the first aircraft, wherein each flight of the first group of flights occurred before the heavy maintenance of the first aircraft and wherein, during each flight of the first group of flights, the first aircraft experienced an overstress event;
displaying the qualification dataset; and
receiving a selection of the fatigue-related parameters;
for a second aircraft of the fleet that experienced heavy maintenance without replacement of the structural component of the second aircraft:
calculating a verification dataset of the selection of fatigue-related parameters from data collected during a second group of flights of the second aircraft, wherein each flight of the second group of flights occurred before the heavy maintenance of the second aircraft and wherein, during each flight of the second group of flights, the second aircraft experienced an overstress event;
displaying the verification dataset; and
receiving a set of the selection of the fatigue-related parameters, wherein the set of the selection of the fatigue-related parameters is less than all of the fatigue-related parameters;
calculating a training dataset of the set of the selection of the fatigue-related parameters from data collected during additional flights of respective aircraft of the fleet, wherein each of the respective aircraft experienced heavy maintenance, wherein each flight of the additional flights occurred before the heavy maintenance of the respective aircraft of that flight, and wherein each flight of the additional flights is a flight during which the respective aircraft experienced an overstress event; and training a predictive model for structural repair during heavy maintenance with the training dataset.

2 . The method of claim 1 , wherein the overstress events of the flights of the first group of flights, the overstress events of the flights of the second group of flights, and the overstress events of the flights of the additional flights of the respective aircraft are each independently selected from the group consisting of a hard landing, a positive acceleration above a predetermined positive acceleration threshold, and a negative acceleration below a predetermined negative acceleration threshold.

3 . The method of claim 1 , wherein the fatigue-related parameters include at least one of a strain of an aerodynamic structure, a difference of strains, an acceleration, a pitch rate, a roll rate, a yaw rate, and a speed brake deployment event.

4 . The method of claim 1 , wherein the structural component of the first aircraft and the structural component of the second aircraft are the same type of structural component and selected from the group consisting of a frame member, a longeron, a stringer, a former, a strut, a beam, a web, a support, a linkage, a splice, and a panel.

5 . The method of claim 1 , wherein the displaying the qualification dataset includes visualizing the qualification dataset and wherein the displaying the verification dataset includes visualizing the verification dataset.

6 . The method of claim 1 , wherein the displaying the qualification dataset includes displaying responsive to user inputs and wherein the displaying the verification dataset includes displaying responsive to user inputs.

7 . The method of claim 1 , wherein training the predictive model includes applying machine learning to the training dataset.

8 . The method of claim 1 , wherein the selection of the fatigue-related parameters is less than all fatigue-related parameters.

9 . The method of claim 1 , wherein the set of the selection of the fatigue-related parameters is less than all of the selection of the fatigue-related parameters.

10 . The method of claim 1 , wherein the receiving the selection of the fatigue-related parameters includes receiving, from a user, the selection of the fatigue-related parameters based on the user's determination of a correlation between the replacement of the structural component of the first aircraft during heavy maintenance and the fatigue-related parameters, and wherein the receiving the set of the selection of the fatigue-related parameters includes receiving, from the user, the set of the selection of the fatigue-related parameters based on the user's determination of a lack of correlation between the selection of the fatigue-related parameters of the qualification dataset and the corresponding fatigue-related parameters of the verification dataset.

11 . The method of claim 1 , wherein the calculating the qualification dataset includes restricting the qualification dataset to the fatigue-related parameters from data collected during the overstress event of at least one flight of the first group of flights of the first aircraft, wherein the calculating the verification dataset includes restricting the verification dataset to the fatigue-related parameters from data collected during the overstress event of at least one flight of the second group of flights of the second aircraft, and wherein the calculating the training dataset includes restricting the training dataset to the fatigue-related parameters from data collected during the overstress event of at least one flight of the additional flights of the respective aircraft.

12 . The method of claim 1 , further comprising collecting flight data by flying a flight of a subject aircraft, and calculating a structural component replacement prediction of a subject structural component of the subject aircraft based on the predictive model and the flight data collected from the flight of the subject aircraft.

13 . The method of claim 1 , further comprising deploying the predictive model for structural repair during heavy maintenance to the fleet of aircraft.

14 . A hybrid feature selection system to generate a predictive model for structural repair during heavy maintenance in a fleet of aircraft that each includes a structural component, the hybrid feature selection system comprising:
a flight database of data collected during flights of aircraft of the fleet of aircraft; a qualification module configured to retrieve qualification data from the flight database, configured to calculate a qualification dataset by applying fatigue-related parameters to the qualification data, configured to display the qualification dataset, and configured to receive, from a user, a selection of the fatigue-related parameters, wherein the qualification data is data collected during a first group of flights of a first aircraft of the fleet that experienced a replacement of the structural component during heavy maintenance of the first aircraft, wherein each flight of the first group of flights occurred before the heavy maintenance of the first aircraft, and wherein, during each flight of the first group of flights, the first aircraft experienced an overstress event; a verification module configured to retrieve verification data from the flight database, configured to calculate a verification dataset by applying the selection of the fatigue-related parameters to the verification data, configured to display the verification dataset, and configured to receive, from the user, a set of the selection of the fatigue-related parameters that is less than all

Assignees

• Boeing Co

Inventors

• Ethington James M
• Sturlaugson Liessman

Classifications

• G06F30/15 Primary

  Vehicle, aircraft or watercraft design · CPC title

• G06Q10/20

  Administration of product repair or maintenance · 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

• G06N5/022

  Knowledge engineering; Knowledge acquisition · CPC title

• G06F17/5095 Primary

  Physics · mapped topic