Predictive Maintenance For Aircraft Engines Using Data Fusion

Keywords

Degradation prognostics; Multiple sensor data fusion; Random forests; Spectral analysis

Abstract

The airline industry spent over $60 billion on maintenance, repair and overhaul of aircraft engines in 2014. This cost is estimated to reach $90 billion in 2024. Many believe that effective prognostics and health monitoring (PHM) systems for aircraft engines will significantly reduce maintenance costs as well as increase the remaining useful life (RUL) of aircraft engines. While in general, model-based prognostic approaches have been demonstrated for damage propagation prediction, little research has been reported on the effectiveness of data-driven prognostics for aircraft engines. This paper presents a new methodology that estimates the RUL of an aircraft engine using multiple sensors and random forests. This new method is demonstrated on a dataset generated by the commercial modular aero-propulsion system simulation (C-MAPSS). Experimental results have shown that a relative error rate of 0.39% can be achieved.

Publication Date

1-1-2018

Publication Title

IISE Annual Conference and Expo 2018

Number of Pages

895-900

Document Type

Article; Proceedings Paper

Personal Identifier

scopus

Socpus ID

85054029299 (Scopus)

Source API URL

https://api.elsevier.com/content/abstract/scopus_id/85054029299

This document is currently not available here.

Share

COinS