Attention and Long Short-Term Memory Network for Remaining Useful Lifetime Predictions of Turbofan Engine Degradation

Machine Prognostics and Health Management (PHM) is often concerned with the prediction of Remaining Useful Lifetime (RUL) assets. Accurate real-time RUL predictions enable equipment health assessment maintenance planning. In this work, we propose a Long Short-Term Memory (LSTM) network combined global Attention mechanisms to learn relationships directly from time-series sensor data. We use NASA Commercial Modular Aero- Propulsion System Simulation (C-MAPPS) datasets assess performance our proposed method. compare approach current state-of-the-art methods on same show that results yield competitive results. Moreover, method does not require previous degradation knowledge, attention weights can be used visualise temporal between inputs predicted outputs.