Energy-efficient operation of neural operators for virtual sensing

arXiv:2609.30580v1 Announce Type: new Abstract: Virtual sensing repeatedly reconstructs physical fields from changing observations, often on a fixed geometry. We investigate how shared spatial computation reduces the energy of these updates while retaining the selected checkpoint and its evaluated predictions. In a heat-exchanger service, standard compiler freezing and explicit trunk reuse give similar operating energy reductions relative to graph replay: approximately 1% at one request per second and 20% at forty requests per second. In 15 W mode with fixed clocks, reuse with graph replay completes the same request sequence with 22.0 to 22.5% less energy than eager execution, including preparation and waiting. DeepONet and Fourier neural operator (FNO) controls distinguish the effects of reusable arithmetic and launch overhead. Preparation, artifact construction, and worker replacement add costs outside repeated inference. These results connect operator structure to operating energy and show how update frequency and execution lifetime govern the benefit of computation reuse in physical-field virtual sensing.

Related

• Neuroscience Inspired Graph Operators Towards Edge-Deployable Virtual Sensing for Irregular Geometries
• Leveraging Differentiable PDE Solvers for Semi-Neural Spatial Reconstruction From Sparse Measurements
• Sensitivity-Constrained Neural Operators for Data-Efficient Forward and Inverse Modeling of Partial Differential Equation Systems
• Efficient Approximation for Encoder--Decoder Neural Operators via Variation Spaces

Source: arXiv cs.LG | 2026-09-28