Neuroscience Inspired Graph Operators Towards Edge-Deployable Virtual Sensing for Irregular Geometries

TL;DR

This paper introduces VS-GNO, a neural operator leveraging spectral-spatial convolution and variable spiking neurons, enabling energy-efficient, real-time virtual sensing on edge devices for complex geometries.

Contribution

The development of VS-GNO, integrating spectral-spatial analysis with variable spiking neurons and an energy-error loss, addressing power constraints in edge virtual sensing.

Findings

Achieved 0.71% reconstruction error with 15% spiking in spectral-only form.

Achieved 1.04% error with 24.5% spiking in full form.

Demonstrated potential for energy-efficient, real-time virtual sensing in irregular geometries.

Abstract

Predicting full-field physics through the real-time virtual sensing of engineering systems can enhance limited physical sensors but often requires sparse-to-dense reconstruction, complex multiphysics, and highly irregular geometries as well as strict latency and energy constraints for edge-deployability. Neural operators have been presented as a potential candidate for such applications but few architectures exist that explicitly address power consumption. Spiking neuron integration can provide a potential solution when integrated on neuromorphic hardware but the current existing neuron models result in severe performance degradation towards regression-based virtual sensing. To address the performance concerns and edge-constraints, we present the Variable Spiking Graph Neural Operator (VS-GNO) which integrates a sophisticated spectral-spatial convolutional analysis and a previously developed Variable Spiking Neuron (VSN) and energy-error balance loss function. With a non-spiking $L_2$ error baseline of $0.4\%$, VS-GNO can provide a reconstruction error of $0.71\%$ with $15\%$ average spiking in its spectral-only form and $1.04\%$ with $24.5\%$ spiking in its entire form. These results position VS-GNO as a promising step towards energy-efficient, edge-deployable neural operators for real-time sparse-to-dense virtual sensing in complex, highly irregular engineering environments.