Virtual Sensing to Enable Real-Time Monitoring of Inaccessible Locations & Unmeasurable Parameters
arXiv:2412.00107v2 Announce Type: replace-cross Abstract: Real-time monitoring of safety-critical interior states remains an open problem in energy systems where physical instrumentation is infeasible. Existing approaches rely on explicit governing equations, finite-dimensional state vectors, or per-instance retraining, which prevents mesh-independent, field-level inference at arbitrary interior coordinates under real-time constraints. We introduce operator-based virtual sensing for nuclear-grade thermal-fluid systems: we use the neural-operator framework to learn solution operators that map s
The advancement in neural operators and the pressing need for safer and more efficient energy systems, particularly nuclear, drive the development of real-time monitoring solutions that overcome physical instrumentation limitations.
This development allows for improved safety, efficiency, and potentially broader adoption of advanced energy systems by enabling real-time insights into previously unmeasurable internal states.
The ability to accurately monitor inaccessible or unmeasurable parameters in complex systems, such as nuclear reactors, without physical sensors fundamentally changes how these systems can be designed, operated, and maintained.
- · Nuclear energy sector
- · Energy producers
- · AI/ML developers
- · Industrial safety equipment manufacturers
- · Traditional sensor manufacturers (niche applications)
Real-time virtual sensing reduces operational risks and costs in high-stakes industrial environments.
Enhanced safety and efficiency could accelerate the deployment of advanced nuclear reactors and other energy technologies.
The widespread adoption of virtual sensing could lead to the development of fully autonomous, self-optimizing industrial plants, ushering in a new era of industrial automation.
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.AI