Anomaly Detection Using Transient Engine Performance Data, 09-R9776

Printer Friendly Version

Principal Investigator
Matthew B. Ballew

Inclusive Dates:  01/01/08 – Current

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, only steady-state data, such as cruise data, are being used for Engine Trending and Diagnostics (ET&D). High-stress transient conditions (e.g., throttle advance during takeoff) are currently being ignored. This project is investigating the feasibility of analyzing currently ignored transient data because it is thought that impending engine failures can be detected earlier in high stress transient conditions than in steady-state conditions.

Approach - The objective of this project is to determine if better anomaly detection and diagnostic methodologies can be developed using transient engine performance data to better assess the health and condition of expensive assets. The following tasks were performed 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: Develop Reader. A software program had to be altered to extract transient performance data because it was originally designed to ignore transient conditions.
  • Task 3: Develop Method of Retrieving Conditional Data from Database. This task explored various ways to define a transient condition and refined the requirements that filter the data.
  • Task 4: Develop Method to Analyze Transient Data for Anomalies. The objective of this task was to develop a method to analyze the filtered data for anomalies. This included developing a method to present data in a repeatable format to the selected statistical algorithms and to develop those statistical algorithms.
  • Task 5: Compare Anomalies to Engine Maintenance Records. During this task, performance anomalies were detected and then correlated to historic maintenance data to test the accuracy and precision of the algorithms.

Accomplishments - All tasks have either been accomplished or are near completion. Methods of conditioning the nonlinear transient data have provided baselines from which adverse engine conditions are detected. The statistic used in the project proved to be an effective method of identifying anomalies and charting trends. The decomposition methods have proved to be effective in identifying which parameters caused observations to signal. Preliminary results suggest a proof-of-concept and demonstration tool will be ready to present to current customers and potential customers in other industries at the close of the project.

2008 Program Home