RGCNU: Recurrent Graph Convolutional Network With Uncertainty Estimation for Remaining Useful Life Prediction

Dear Editor, This letter focuses on the problem of remaining useful life (RUL) prediction equipment. Existing graph neural network (GCN)-based approaches merely provide point estimation RUL. However, estimated RUL often varies widely due to model parameters and noise in data. It is important know uncertainty predictions for reliable risk analysis maintenance decision making. To map relationship between noisy condition monitoring data with uncertainty, we propose a recurrent convolutional (RGCNU) prediction. In our approach, correlation exploiting module captures spatial-temporal correlations based learned structure. Furthermore, fusion associates improve robustness