Remaining Useful Life Prediction using Dynamic Principal Component Analysis and Deep Gated Recurrent Unit Network

Massive multi-sensor data makes it challenging to extract degradation features and predict the remaining useful life (RUL). This paper introduces a novel RUL prediction method using dynamic principal component analysis (DPCA) and deep gated recurrent unit (DGRU) network. DPCA is utilized to construct the health index (HI) curve considering autocorrelation. DGRU network is built to predict the RUL based on the constructed HI curves. The dataset of turbofan engines is used to verify the proposed method. The results show that the proposed RUL framework creates a higher accuracy than other related work.