Quantum computing-enhanced large-scale residential electric vehicle charging management

This paper presents an innovative quantum computing framework for residential EV charging management, leveraging smart meter data and advanced quantum optimization techniques under time-of-use pricing schemes. Our approach began with the extraction of household electricity usage patterns and occupancy profiles, emphasizing EV charging behaviors. By integrating occupancy patterns, we determined charging feasibility and defined operational constraints. Generative Adversarial Network was employed for additional load profiles for large-scale simulation. The EV charging states were represented as binary variables and exploiting qubit superposition. The optimal control problem aimed at minimizing daily electricity costs was formulated with EV charging and discharging actions subject to battery capacity and occupancy constraints. This problem was reformulated as a quadratic unconstrained binary optimization model through the inclusion of penalty functions. We implemented D-Wave quantum computer to solve. Simulation studies conducted on both single user and community level with 225 and 1000 households revealed significant performance improvements. Quantum computing achieved peak load reductions of up to 94.2 % and average daily electricity bill savings of approximately 34.7 %, with computing times ranging from seconds to minutes. Comparison with conventional solvers including Gurobi and Matlab demonstrated that quantum computing was particularly advantageous for discrete EV charging optimization. While continuous methods provided marginally higher precision, the quantum approach offered superior scalability and efficiency for large-scale problems. These results highlight the transformative potential of quantum computing for real-time, adaptive EV management in smart grids.