🤖 AI Summary
This study addresses the limitations of conventional computing reliant on specialized hardware by introducing, for the first time, a “power-grid-native computing” paradigm that leverages existing electrical grids as fixed physical computational operators. Methodologically, computation is executed through voltage perturbations and current responses, integrating time-domain simulation, Kirchhoff’s laws, power electronic interfaces, and surrogate models to construct a comprehensive encoding–decoding framework. This approach substantially reduces trainable parameters while supporting spatial concurrency and temporal multiplexing. Experimental evaluations demonstrate classification accuracies of 91.5% and 82.25% on MNIST and Fashion-MNIST, respectively. These results validate the critical influence of grid topology and signal representation on computational utility, establishing a promising foundation for physics-based analog computing.
📝 Abstract
Computing is conventionally implemented by hardware engineered for information processing. Here we investigate infrastructure-native computing: the use of a physical system built for another primary function as a fixed computational operator. In time-domain simulations of an IEEE 14-bus electrical network, Kirchhoff's current law and Ohm's law relate voltage-reference perturbations applied at distributed controllable nodes interfaced by power electronics converters to current responses through a topology-dependent transformation. A trained digital encoder and decoder exploit this transformation for image classification, reaching 91.5% accuracy on MNIST and 82.25% on Fashion-MNIST. The modeled operator is represented by 933 surrogate parameters, compared with 12,340 task-trained parameters for an accuracy-matched fully connected core transformation. Current superposition further supports concurrent spatial sharing of the operator and sequential temporal reuse, with per-stream accuracies above 85% and 93%, respectively, in surrogate-model evaluations. Evaluations on CIFAR-10 and repeated 10-class tasks sampled from a Butterflies-and-Moths dataset show that the incremental utility of the physical operator depends on the representation supplied by upstream digital feature extraction. These results provide a simulation-based proof of concept for infrastructure-native computing with electrical networks and identify topology, accessible control channels, and input representation as determinants of its computational utility.