Turbofan Engine Health Degradation Prediction Project
The aim of this project is to apply a data-driven approach to produce a model able to accurately predict the rest of useful life remaining (RUL) of a jet engine based on a historical dataset. Specifically, we are using an open dataset N_CMAPSS, from NASA (Chao et al., 2021) containing system sensor readings and failure modes for a number of Turbofan Engines. This project is in collaboration with BAE SYSTEMS, an aerospace company that supports customers across the whole life cycle of the air sector and shows an interest in developing data-driven algorithms for Integrated Vehicle Health Management (IVHM).
The prediction is centred around the RUL; an estimation of how many flight cycles are remaining before a component requires replacement (Kang et al., 2021). To provide a comparable measurement of the performance of the model, we will utilise the root mean square error (RMSE), given by the equation:
This Github repo contains all contains all code involved in the project.