Power network optimization: a quantum approach

TL;DR

This paper demonstrates how quantum annealing can optimize electricity surplus in power networks more effectively than classical methods, using hybrid quantum-classical approaches tested on D-Wave and Azure Quantum platforms.

Contribution

It introduces a quantum approach to power network optimization, formulating the problem as a QUBO and comparing quantum and classical solutions.

Findings

Hybrid quantum approaches outperform classical methods in solution quality.

Quantum solutions consistently achieve lower objective function values.

The approach is tested on various problem sizes with positive results.

Abstract

Optimization of electricity surplus is a crucial element for transmission power networks to reduce costs and efficiently use the available electricity across the network. In this paper we showed how to optimize such a network with quantum annealing. First, we define the QUBO problem for the partitioning of the network, and test the implementation on purely quantum and hybrid architectures. We then solve the problem on the D-Wave hybrid CQM and BQM solvers, as well as on classical solvers available on Azure Quantum cloud. Finally, we show that the hybrid approaches overperform the classical methods in terms of quality of the solution, as the value of the objective function of the quantum solutions is found to be always lower than with the classical approaches across a set of different problem size.