11 - Operations and Sustainment
TOWARDS TRUSTWORTHY DATA-DRIVEN GAS TURBINE PROGNOSTICS
A. Apostolidis¹, S. Le Dantec, SIGMA Clermont, France; K.P. Stamoulis¹; ¹Amsterdam University of Applied Sciences, Netherlands
This paper discusses the evolution of aircraft data recording, focusing on Aircraft Operational Data for prognostics. The importance of trustworthiness for data-driven prognostics is pivotal. This work uses real-world engine data for Exhaust Gas Temperature prediction, addressing challenges like missing data points, and explores solutions like KNN-imputation and Generalized Additive Model.