Quantum computing-based solver for interacting power grids

Abstract

The proliferation of power electronics in multi-terminal transmission grids has increasingly led to harmonic distortions and dynamic instabilities. While Resonance Mode Analysis (RMA) provides deep insights into these system resonances, evaluating the critical modes of large-scale grids presents a severe computational bottleneck. Classical iterative techniques must continuously diagonalize massively high-dimensional, non-Hermitian admittance matrices across a wide frequency spectrum, a process that rapidly exhausts classical memory and processing limits. To overcome this scaling barrier, we propose a novel quantum-classical hybrid methodology that natively maps the transmission grid's admittance matrix onto a Quantum Processing Unit (QPU). Because the grid's matrix is non-Hermitian, standard quantum eigensolvers are insufficient; thus, we employ the Real Variance-based Variational Quantum Eigensolver (RVVQE) algorithm to accurately extract the complex eigenvalues that represent the system's modes. Validated against a standard 5-bus transmission system, the quantum-derived critical-resonance modal impedances demonstrate near-perfect alignment with the exact classical frequency responses. Crucially, by encoding the grid's state logarithmically into quantum memory, this methodology bypasses classical RAM limitations. The successful implementation of the RVVQE framework not only bridges the mathematical topologies of dissipative electrical grids and open quantum systems but also provides a profoundly scalable architecture capable of diagnosing resonance instabilities in massive, continental-scale networks that currently exceed classical computational boundaries.