Transactions on Machine Learning and Data Mining (ISSN: 1865-6781)
Volume 7 - Number 2 - October 2014 - Pages 41-63
Interpretable Aircraft Engine Diagnostic via Expert Indicator Aggregation"Tsirizo Rabenoro*, Jerome Lacaille*, Marie Cottrell**, Fabrice Rossi**
* Snecma, Groupe Safran, Moissy Cramayel, France
**SAMM, Universite Paris 1, France
Detecting early signs of failures (anomalies) in complex systems is one of the main goal of preventive maintenance. It allows in particular to avoid actual failures by (re)scheduling maintenance operations in a way that optimizes maintenance costs. Aircraft engine health monitoring is one representative example of a eld in which anomaly detection is crucial. Manufacturers collect large amount of engine related data during ights which are used, among other applications, to detect anomalies. This arti- cle introduces and studies a generic methodology that allows one to build automatic early signs of anomaly detection in a way that builds upon hu- man expertise and that remains understandable by human operators who make the nal maintenance decision. The main idea of the method is to generate a very large number of binary indicators based on parametric anomaly scores designed by experts, complemented by simple aggregations of those scores. A feature selection method is used to keep only the most discriminant indicators which are used as inputs of a Naive Bayes classi- er. This give an interpretable classier based on interpretable anomaly detectors whose parameters have been optimized indirectly by the selec- tion process. The proposed methodology is evaluated on simulated data designed to reproduce some of the anomaly types observed in real world engines.
Keywords: Engine Health Monitoring; Turbofan; Fusion; Anomaly Detection
Download Paper (348 KB)