Reconfiguración óptima de redes de distribución con generación renovable mediante el algoritmo híbrido cuántico-clásico qaoa para minimización de pérdidas de potencia

Abstract

This work presents a quantum-classical hybrid approach for the optimal reconfiguration of distribution networks with renewable generation, integrating the QAOA algorithm with the nonlinear AC power flow. The methodology transforms the combinatorial problem of 32 configurations of the IEEE 33-bus system into a QUBO formulation of 15 parameters through exhaustive evaluation by Newton-Raphson, maps the coefficients to the Ising Hamiltonian, and executes the variational optimization through a quantum circuit of unit depth. The algorithm identifies in 1.36 seconds the configuration that closes the tie- switches SW34 and SW37, establishing an alternative radial topology that reduces the active losses from 282.938 kW to 95.773 kW, equivalent to 66.15% improvement. The solution presents a deviation of 9.98% with respect to the global optimum, raises the minimum voltage to 0.9531 p.u. complying with the IEEE 1547 normative limits, and redistributes the flows reducing the main current from 246 A to 141 A. The synergy between the distributed generation and the quantum reconfiguration achieves an efficiency of 78.0%, demonstrating that QAOA overcomes the limitations of the premature convergence of the classical metaheuristics in reconfiguration problems with nonlinear constraints.