Among the various data-driven approaches used for RUL prediction, Recurrent Neural Networks (RNNs) have certain prima facie advantages over other approaches because the connections between internal nodes form directed cycles, thus creating internal states which enables the network to encapsulate dynamic temporal behavior and also to properly handle the noise affecting the collected signals. How...
