A remaining useful life prediction method for rotating machinery based on interactive BiLSTM

Abstract Rotating machinery, as a key component of mechanical equipment, plays a crucial role in ensuring the reliability and safety of the equipment. This article proposes a RUL prediction framework based on interactive bidirectional long short-term memory (IT-BiLSTM) and function fitting, aiming to improve the accuracy of RUL prediction in rotating machinery. Firstly, to unify the failure threshold under different working conditions, a data interception method based on root mean square is proposed, and empirical mode decomposition is used to reconstruct the intercepted signals; then, 16 features are extracted, and health indicator (HI) is constructed through IT-BiLSTM, which is mapped to the [0,1] interval; finally, the time when HI is 0 is found through function fitting, which is the current RUL. The accuracy of the proposed method is verified using two datasets. The results show that the proposed IT-BiLSTM can more accurately predict the RUL of rotating machinery.