Incipient Fault Point Detection Based on Multiscale Diversity Entropy

Abstract

If the incipient fault (IF) point is not corrected, it may imply a poor prognostic framework, a false value of remaining useful life (RUL), and unexpected catastrophic failure. The use of the concept of multiscale diversity entropy (MDE) in the context of predicting IF is a novel area that has yet to be fully explored. Since MDE is commonly used for measuring a system’s dynamic complexity from a signal, it is worth exploring for predicting IF. Can MDE be used to develop a framework to predict the IF of a system? This study developed a new framework to determine the IF. The performance of the framework was demonstrated on bearing data and battery data. In addition, the results of this study were compared with another methodology, widely used for predicting IF. In conclusion, this new methodology produces a more accurate prediction of IF because of the physical principles associated with MDE.