Advanced Statistics for Improved and Automated Engine Trending and Diagnostics, 09-R9825
Printer Friendly VersionPrincipal Investigator
Matthew B. Ballew
Inclusive Dates: 05/27/08 – 09/27/08
Background - Jet engine performance data is analyzed for adverse trends or step changes based on an engine's operating baseline. An undetected engine component failure can lead to unscheduled maintenance actions, reduced asset availability, and catastrophic engine failures soon after the initial component failure. Currently, these failures must be detected manually by an engine trender. The purpose of this project was to address issues of undetected failures and other errors introduced by the current Engine Trending and Diagnostics (ET&D) process.
Approach - The objective of this project was to determine if better detection and diagnostic capabilities can be developed and automated using advanced statistical methods to better determine the health and condition of expensive assets. The following tasks were completed to reach this objective:
- Task 1: Gather Data. This task required
gathering all available engine performance data and maintenance data to
support the following tasks.
- Task 2: Apply Statistical Methodologies to
Engine Performance Data. This task explored various statistical algorithms
to apply to engine performance data.
- Task 3: Automate Anomaly Detection. The purpose
of this task was to implement the algorithms from Task 2 into software that
demonstrates the automation and accuracy potentials of the algorithms.
- Task 4: Compare Anomalies to Engine Maintenance Records. During this task, performance anomalies were detected and then correlated to historic maintenance data to test and validate the accuracy and precision of the algorithms.
Accomplishments - All tasks were successfully completed. The statistic used in the project proved to be an effective method of identifying anomalies and charting trends. Project results provided a proof-of-concept and demonstration tool to present to current customers and potential customers in other industries. Figure 1 shows an example of the statistic that immediately detected a variable geometry issue at flight 32 which, using the current ET&D process, was not identified for over 14 months.
