As an important part of prognostics and health management, remaining useful life (RUL) prediction can provide users and managers with system life information and improve the reliability of maintenance systems. Data-driven methods are powerful tools for RUL prediction because of their great modeling abilities. However, most current data-driven studies require large amounts of labeled training data and assume that the training data and test data follow similar distributions. In fact, the collected data are often variable due to different equipment operating conditions, fault modes, and noise distributions. As a result, the assumption that the training data and the test data obey the same distribution may not be valid. In response to the above problems, this paper proposes a data-driven framework with domain adaptability using a bidirectional gated recurrent unit (BGRU). The framework uses a domain-adversarial neural network (DANN) to implement transfer learning (TL) from the source domain to the target domain, which contains only sensor information. To verify the effectiveness of the proposed method, we analyze the IEEE PHM 2012 Challenge datasets and use them for verification. The experimental results show that the generalization ability of the model is effectively improved through the domain adaptation approach.

Predicting remaining useful life (RUL) is a crucial part of prognostics and health management (PHM), which has attracted widespread attention in academia and industry over the past few decades. Effective estimation of RUL is predicated on the development of a suitable degradation model for the system. However, most of the existing models require offline learning of a priori parameters, which is not applicable in the absence of historical data. To solve this problem, a stochastic degradation model based on the Lévy process is proposed in this study. The utilization of the Kalman filter and expectation–maximization in conjunction with Rauch–Tung–Striebel (KF-EM-RTS) enables the acquisition and real-time updating of model parameters and the probability density function (pdf) for RUL. This approach also allows for the characterization of parameter uncertainty. Simultaneously, the analytical expression of the RUL pdf can be obtained by the definition based on first hitting time (FHT) and Lévy–Khinchin formula. The high-speed computer numerical control (CNC) milling machine cutters dataset from the 2010 IEEE Data Challenge and solid-state RF power amplifier (SSRFPA) degradation data are utilized to validate the proposed model. The results of the experiments demonstrate that the proposed model is capable of delivering accurate RUL estimations.

Forecasting the remaining useful life (RUL) is a crucial aspect of prognostics and health management (PHM), which has garnered significant attention in academic and industrial domains in recent decades. The accurate prediction of RUL relies on the creation of an appropriate degradation model for the system. In this paper, a general representation of diffusion process models with three sources of uncertainty for RUL estimation is constructed. According to time-space transformation, the analytic equations that approximate the RUL probability distribution function (PDF) are inferred. The results demonstrate that the proposed model is more general, covering several existing simplified cases. The parameters of the model are then calculated utilizing an adaptive technique based on the Kalman filter and expectation maximization with Rauch-Tung-Striebel (KF-EM-RTS). KF-EM-RTS can adaptively estimate and update unknown parameters, overcoming the limits of strong Markovian nature of diffusion model. Linear and nonlinear degradation datasets from real working environments are used to validate the proposed model. The experiments indicate that the proposed model can achieve accurate RUL estimation results.

As a critical part of prognostics and health management (PHM), remaining useful life (RUL) prediction can provide manufacturers and users with system lifetime information and improve the reliability of maintainable systems. Particle filters (PFs) are powerful tools for RUL prediction because they can represent the uncertainty of results well. However, due to the lack of measurement data, the parameters of the measurement model cannot be updated during the long-term prediction process. Additionally, for complex systems, the measurement model of a system often cannot be obtained in an analytical form. In this paper, a fusion prognostic method based on an extended belief rule base (EBRB) and a PF is designed to solve these problems. In the proposed framework, a double-layer maximum mean discrepancy-extended belief rule base (DMMD-EBRB) model with time delay is adopted to estimate and predict the hidden behavior of a degrading system. The unknown parameters of the degradation model are identified by the PF using the output of the EBRB. Afterwards, the system state is further predicted by the PF. The effectiveness of the proposed method is validated with the NASA-PCoE and CALCE lithium-ion battery degradation experiment datasets. In addition, several other related fusion methods are investigated for comparison with the proposed method. The experiments show that the proposed method yields better performance than the existing methods.