Real time streaming analytics for flight data processing

Patent metadata

Field	Value
Publication number	US-10672204-B2
Application number	US-201715813428-A
Country	US
Kind code	B2
Filing date	Nov 15, 2017
Priority date	Nov 15, 2017
Publication date	Jun 2, 2020
Grant date	Jun 2, 2020

Abstract

An example method includes receiving a plurality of data streams acquired for a respective parameter of a plurality of parameters indicating an operating condition of the aircraft, selecting at least one data stream corresponding to at least one parameter of the respective plurality of parameters, selecting a portion of data from the at least one data stream, comparing the portion of data to a model determined for the at least one parameter based on historical data, determining that a failure has occurred or is likely to occur during operation of the aircraft based on the comparing, and transmitting aircraft health monitoring information indicative of occurrence or likelihood of occurrence of the failure.

First claim

What is claimed is:

1. A method implemented by a computing device for data stream analytics for an aircraft, the method comprising:
receiving a plurality of data streams acquired for a respective parameter of a plurality of parameters indicating an operating condition of the aircraft, wherein the aircraft is configured to operate in different flight modes, each flight mode being associated with a respective set of operating rules;
selecting, from the plurality of data streams, at least one data stream corresponding to at least one parameter of the plurality of parameters;
selecting a portion of data from the at least one data stream based on which flight mode the aircraft is operating in when the portion of data is acquired;
comparing the portion of data to a model determined for the at least one parameter based on historical data;
determining that a failure has occurred or is likely to occur during operation of the aircraft based on the comparing; and
transmitting aircraft health monitoring information indicative of occurrence or likelihood of occurrence of the failure.

2. The method of claim 1 , wherein the plurality of data streams are acquired during different flights of the aircraft or different flight segments.

3. The method of claim 1 , wherein the at least one data stream includes gaps between data points of the at least one data stream, and wherein the method further comprises:
estimating missing data points in the gaps of the at least one data stream.

4. The method of claim 1 , wherein the at least one data stream is constructed from data segments obtained from multiple respective sources, the method further comprising:
reordering the data segments such that the at least one data stream represents variation of the at least one parameter in a time-based sequential order.

5. The method of claim 1 , further comprising:
identifying a sequence of events occurring on the aircraft based on the portion of data; and
determining that the sequence of events is out of order compared to an expected order, wherein determining that the failure has occurred or is likely to occur is based on determining that the sequence of events is out of order.

6. The method of claim 1 , further comprising:
determining a statistical variable for the at least one parameter, wherein comparing the portion of data to the model determined for the at least one parameter based on the historical data comprises: comparing the determined statistical variable to a stored statistical variable associated with the historical data.

7. The method of claim 6 , wherein the statistical variable is one statistical variable among a group of statistical variables determined using the portion of data, wherein comparing the portion of data to the model comprises comparing the statistical variables to respective threshold values associated with the historical data, and wherein determining that the failure has occurred or is likely to occur is based on a majority of the statistical variables meeting the respective threshold values.

8. The method of claim 1 , wherein receiving the plurality of data streams comprises receiving raw time-spaced data points associated with sensor measurement of the at least one parameter from a flight data warehouse coupled to the aircraft, the method further comprising:
determining variation of the sensor measurement of the at least one parameter over time.

9. The method of claim 1 , further comprising:
detecting, using the at least one data stream, an event that has occurred on the aircraft without being initiated by a pilot, wherein selecting the portion of data comprises selecting the portion of data that followed the event in the at least one data stream.

10. The method of claim 1 , wherein the different flight modes comprise pre-flight, engine taxi out, take-off, climb-out, cruise, descent, approach, landing, taxi-in, engine shut-down, and post flight.

11. A flight data processing system comprising:
a flight data conversion computing device having one or more processors configured to (i) receive, from a flight data warehouse, raw time-space data points of a plurality of data streams acquired for a respective parameter of a plurality of parameters indicating an operating condition of an aircraft, wherein the aircraft is configured to operate in different flight modes, each flight mode being associated with a respective set of operating rules, and (ii) convert the raw time-spaced data points of the plurality of data streams into measurements representing variation of the plurality of parameters over time;
a database in communication with the flight data conversion computing device and configured to store converted plurality of data streams generated by the flight data conversion computing device; and
a streaming analytics module in communication with the flight data conversion computing device or the database, wherein the streaming analytics module comprises:
one or more processors, and
memory storing thereon instructions, that when executed by the one or more processors, cause the streaming analytics module to perform operations comprising:
selecting, from the converted plurality of data streams, at least one data stream corresponding to at least one parameter of the plurality of parameters,
selecting a portion of data from the at least one data stream based on which flight mode the aircraft is operating in when the portion of data is acquired,
comparing the portion of data to a model determined for the at least one parameter based on historical data,
determining that a failure has occurred or is likely to occur during operation of the aircraft based on the comparing, and
transmitting aircraft health monitoring information indicative of occurrence or likelihood of occurrence of the failure.

12. The flight data processing system of claim 11 , wherein the database is configured to store the converted plurality of data streams in a Hadoop Distributed File System to facilitate parallel processing of the converted plurality of data streams by the streaming analytics module.

13. The flight data processing system of claim 11 , further comprising:
a historical data processing computing device in communication with the database and configured to determine the model based on historical converted plurality of data streams received from the database.

14. The flight data processing system of claim 11 , wherein the operations further comprise:
determining a statistical variable for the at least one parameter, wherein comparing the portion of data to the model determined for the at least one parameter based on the historical data comprises: comparing the determined statistical variable to a stored statistical variable associated with the historical data.

15. The flight data processing system of claim 14 , wherein the statistical variable is one statistical variable among a group of statistical variables determined using the portion of data, wherein comparing the portion of data to the model comprises comparing the determined statistical variables to respective threshold values associated with the historical data, and wherein determining that the failure has occurred or is likely to occur is based on a majority of the statistical variables meeting the respective threshold values.

16. The flight data processing system of claim 11 , wherein the operations further comprise:
detecting, using the at least one data stream, an event that occurred on the aircraft without being initiated by a pilot, wherein selecting the portion of data comprises selecting the portion of data that followed the event in

Assignees

• Boeing Co

Inventors

• Chopra Seema
• Palyam Sudheer
• Schimert James
• Wilmering Timothy Joseph
• Thumati Balaje

Classifications

• G07C5/0825

  using optical means · CPC title

• G07C5/085

  using electronic data carriers · CPC title

• H04L65/60

  Network streaming of media packets · CPC title

• G05B2219/45071

  Aircraft, airplane, ship cleaning manipulator, paint stripping · CPC title

• G07C5/0808 Primary

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