Detection of Anomalous Behaviour of Aircraft Engines for Predictive Maintenance

Document Type

Conference Article

Publication Title

2023 7th International Conference on System Reliability and Safety, ICSRS 2023

Abstract

This paper presents a methodology for anomaly detection in aircraft engines on the basis of a survival data formulation. A decision threshold is used for the estimated hazard rate under the assumption of the Cox regression model. Utilization of information relating to aging of the engines gives this formulation a natural advantage over the standard methods of anomaly detection through classification or clustering. There is also the scope of taking into account engine heterogeneity through frailty modeling. The proposed method is found to have superior performance over the standard methods for a data set involving hundreds of flights of about a thousand aircraft engines.

First Page

423

Last Page

427

DOI

10.1109/ICSRS59833.2023.10381239

Publication Date

1-1-2023

This document is currently not available here.

Share

COinS