Machine Learning Regression Models for Turbofan Engines: A Comparative Study on Remaining Useful Life Prediction

This study conducts a systematic comparison of data-driven regression models for predicting the Remaining Useful Life (RUL) of turbofan engines in aviation Prognostics Health Management (PHM) field, utilizing the C-MAPSS dataset with advanced feature engineering, the paper evaluates models from simple linear to complex high-dimensional ensemble structures. Notably, the CatBoost model and other ensemble techniques demonstrated superior performance, with CatBoost achieving an R2 value of 99% and other ensembles averaging 92%, significantly outperforming other models with 65% to 77% values. This indicates the robustness of ensemble machine learning methods in precise RUL estimation.