BEARING EARLY FAULT DETECTION AND REMAINING USEFUL LIFE ESTIMATION WITH KNOWLEDGE-INFORMED MACHINE LEARNING
A knowledge-informed warm-start autoencoder that enables efficient adaptation of a pre-trained model to a new bearing with improved accuracy and fair computational effort is proposed, focusing on early fault detection and remaining useful life (RUL) estimation.
Jing-Yi Yan, M. Gardner, Yanwen Xu et al.
· 0 citations