Accurate glucose prediction plays an important role in glucose management and closed-loop insulin delivery for subjects with diabetes. Due to its powerful data mining capability, neural networks are used to grasp the glucose trends from continuous glucose monitoring (CGM) data. However, this approach requires a large number of individual data and plentiful computing resources, but has relatively poor interpretability. Given that the glucose metabolism mechanism model contains abundant physiological information, fusing the information with neural networks can reduce the demand for data and computing resources, and improve interpretability. In this study, a physics-informed glucose-insulin neural network (PIGNN) model is proposed, of which the structure and loss function are designed based on the glucose-insulin dynamic model. According to the experiments of 22 real subjects and 30 in-silico subjects with type 1 diabetes, the prediction accuracy of this method achieves 0.726 ± 0.126 mmol/L. Compared with models without physical information, the proposed PIGNN shows a significant improvement in glucose prediction for real subjects with a limited sample size (only 48 data samples), resulting in an accuracy improvement of 12.33%. In addition, it is proved that with limited data more physical information can improve the glucose prediction accuracy significantly.

Due to the considerable randomness in contaminant size and arrival, as well as unavoidable epistemic uncertainties and manufacturing/installation variations, both parametric and model uncertainties in practice pose a challenge to the oil filter's remaining useful life prediction. The oil filter is a crucial component in guaranteeing hydraulic fluid cleanliness. This paper introduces a remaining useful life prediction method that fuses a physics-based degradation model with data-driven stochastic process models. Based on Bayesian inference, the method achieves effective RUL prediction for oil filters by synthesizing real-time degradation observations with the respective advantages of different candidate models. In comparison to the Ergun and Wiener models, experimental validation shows that the root mean square error（RMSE）of real-time degradation prediction for oil filter achieves 0.003 9 MPa, resulting in drops of 79.9% and 77.5%, respectively. The results indicate that the proposed method exhibits superior prediction accuracy and strong generalization capability, highlighting its practical value for engineering applications.