A Streaming Sparse Cholesky Method for Derivative-Informed Gaussian Process Surrogates Within Digital Twin Applications
arXiv:2511.00366v2 Announce Type: replace-cross Abstract: Digital twins are developed to model the behavior of a specific physical asset (or twin), and they can consist of high-fidelity physics-based models or surrogates. A highly accurate surrogate is often preferred over multi-physics models as they enable forecasting the physical twin future state in real-time. To adapt to a specific physical twin, the digital twin model must be updated using in-service data from that physical twin. In this paper, we combine and extend several previous surrogate-related advancements with the goal of demonst
The increasing complexity and demand for real-time adaptability in industrial and engineering systems necessitate more sophisticated and efficient surrogate modeling techniques.
This development allows for faster, more accurate, and adaptable digital twins, significantly improving predictive maintenance, operational efficiency, and design optimization across various industries.
Digital twins can now incorporate derivative information more effectively for faster model updates and higher fidelity predictions in dynamic environments.
- · Manufacturing sector
- · Aerospace and defense
- · Energy sector
- · AI/ML model developers
- · Traditional physics-based simulation companies (if they don't adapt)
Improved performance and reliability of complex physical assets through real-time predictive capabilities.
Reduced operational costs and downtime due to enhanced fault prediction and preventative maintenance triggered by more accurate digital twin models.
Acceleration of 'lights-out' operations and fully autonomous industrial processes as digital twins achieve near-perfect real-time fidelity, requiring minimal human intervention.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG