Data-Driven and Physics-Based Approaches for Wind Turbine High-Speed Shaft Bearing Prognostics

Remaining useful life (RUL) prediction is a crucial mission for the Prognostic and Health Management (PHM) of machinery equipment in modern industry. In this paper, we use two different methods to predict degradation occurring in high-speed shaft bearing (HSSB) and predict the RUL using Artificial Neural Network (ANN) and Kalman Smoother (KS). To confirm the accuracy and validity of the method, experiments are conducted on a real HSSB dataset. The comparison with current state-of-the-art methods on the same dataset shows the performance of the presented methods.