Neuromorphic Parameter Estimation for Power Converter Health Monitoring Using Spiking Neural Networks

TL;DR

This paper introduces a neuromorphic spiking neural network approach for power converter health monitoring that achieves high accuracy and energy efficiency suitable for edge deployment.

Contribution

It separates spiking temporal processing from physics enforcement, enabling physics-consistent training and improved fault detection in power converters.

Findings

Reduces lumped resistance error from 25.8% to 10.2%.

Achieves approximately 270x energy reduction on neuromorphic hardware.

Detects faults with a 5.5 percentage-point spike-rate increase.

Abstract

Always-on converter health monitoring demands sub-mW edge inference, a regime inaccessible to GPU-based physics-informed neural networks. This work separates spiking temporal processing from physics enforcement: a three-layer leaky integrate-and-fire SNN estimates passive component parameters while a differentiable ODE solver provides physics-consistent training by decoupling the ODE physics loss from the unrolled spiking loop. On an EMI-corrupted synchronous buck converter benchmark, the SNN reduces lumped resistance error from $25.8\%$ to $10.2\%$ versus a feedforward baseline, within the $\pm 10\%$ manufacturing tolerance of passive components, at a projected ${\sim}270\times$ energy reduction on neuromorphic hardware. Persistent membrane states further enable degradation tracking and event-driven fault detection via a $+5.5$ percentage-point spike-rate jump at abrupt faults. With $93\%$ spike sparsity, the architecture is suited for always-on deployment on Intel Loihi 2 or BrainChip Akida.