Real-time adaptation for time-series signal prediction using label-aware neural processes

Building a predictive model that rapidly adapts to real-time condition monitoring (CM) time-series data is critical for engineering systems/units. Unfortunately, many current methods suffer from a trade-off between representation power and agility in online settings. In this paper, we propose a neural process-based approach that addresses this trade-off. It encodes available observations within a CM signal into a representation space and then reconstructs the signal’s history and evolution for prediction. Once trained, the model can encode an arbitrary number of observations without requiring retraining, enabling on-the-spot real-time predictions along with quantified uncertainty and can be readily updated as more online data is gathered. Furthermore, our model is designed to incorporate partial information on qualitative factors (e.g., missing labels) from individual units. This integration not only enhances individualized predictions for each unit but also enables joint inference for both signals and their associated labels. Numerical studies on both synthetic and real-world data in degradation modeling highlight the advantageous features of our model in real-time adaptation, enhanced signal prediction with uncertainty quantification, and joint prediction for labels and signals.