Cheng Jiao
Tonny Liu


Techniques are described for determining a remaining useful life (RUL) prognosis of bearings using a feature extraction module for extracting time series normalized similarity (TSNS) features for vibration data normalization and a prediction module utilizing a deep learning model, known as an independently recurrent neural network (IndRNN), for predicting bearing RUL. The feature extraction module and prediction module are deployed on a fog computing platform as services for determining the RUL prognosis of bearings.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.