This content is not included in your SAE MOBILUS subscription, or you are not logged in.

Dynamic Alert Generation Technology for Health & Usage Monitoring Systems

Journal Article
ISSN: 1946-3855, e-ISSN: 1946-3901
Published October 18, 2011 by SAE International in United States
Dynamic Alert Generation Technology for Health & Usage Monitoring Systems
Citation: Pipe, K. and Culkin, B., "Dynamic Alert Generation Technology for Health & Usage Monitoring Systems," SAE Int. J. Aerosp. 4(2):1197-1206, 2011,
Language: English


Current alert setting methodologies based on setting defect detection alert thresholds for vibration and other Health & Usage Monitoring Systems (HUMS) indicators have many limitations, principally there is necessarily a compromise to be achieved between the true negative and false positive diagnostic metrics. This is true for all alert detection techniques from fixed thresholds through to Support Vector Machines. In this paper we describe techniques, validated using helicopter HUMS data, which do not invoke this compromise and independently minimise both the true negative and false positive rates.
This paper will also demonstrate how the alert processing can be made more robust and overcome the problems introduced by HUMS data being both non-stationary, non-ergodic i.e. characteristics that change both with time and from platform to platform.
Two techniques utilised in the CFAR-Autotrend proprietary alert detection technology are described; a) The Constant False Alarm Rate technique for setting thresholds based on signal amplitude and detecting level changes b) The Box- Car technique for the automatic detection of local trends. These techniques produce alerts that are based on a significant change in local, not global, conditions in the data stream and are not based on an a priori model of the statistics of the signal.
The performance of these techniques on real world HUMS generated vibration data and the engineering of the processes to automatically produce reliable and robust alerts are reviewed. False positives have been verified to be reduced to be less than 10% of true positives; a level where the diagnostics process becomes robust. The sensitivity to defects is improved to a level where the true negative rate has been reduced to a level where the diagnostic processes can be relied on and the prognostic interval becomes reliable. The consequential impact of this level of performance on HUMS dynamic alert processing is assessed.
The automation of alert processing breaks the link between the number of aircraft fitted with HUMS and the manpower required to process the data. More importantly it will permit the scarce resource of HUMS specialists trained to diagnose defects to be concentrated on that task, not dissipated processing false alerts. For the future the Remaining Useful Life (RUL) realised from the increased sensitivity of the technique is at a magnitude where the benefits promised by Condition Based Maintenance can become a reality.