Combined Use of Hybrid Optimization and Quantum-Enhanced Neural Evolution for Predicting Cost and Charge Dynamics of Electric Vehicles

This paper introduces EcoSwarm, a bio-inspired hybrid optimization algorithm aimed at enhancing cost efficiency and charge management in electric vehicle (EV) fleets. EcoSwarm leverages Quantum-Enhanced Neural Evolution (QENE), which integrates quantum computing principles with neural evolution to significantly improve prediction accuracy and operational efficiency. Unlike traditional optimization methods, EcoSwarm employs quantum-inspired probabilistic transitions to explore a broader solution space, increasing the likelihood of achieving global optimal solutions. The proposed approach demonstrates $\mathbf{3 5 . 6 \%}$ reduction in operational costs and achieves $\mathbf{9 2 . 1 \%}$ energy utilization efficiency—surpassing the Particle Swarm Optimization (PSO) and Genetic Algorithm (GA) by at least 11.3% and $15.8 \%$, respectively. Furthermore, EcoSwarm enhances charging demand prediction accuracy by $32 \%$, as measured by the Mean Absolute Error (MAE), showcasing its capability to manage large-scale and dynamic EV fleets effectively. EcoSwarm is highly scalable and adaptable, making it suitable for addressing challenges in large-scale EV fleet management. Future research will explore its application in real-time autonomous fleet management, optimization systems, and smart grids as integral components of sustainable mobility solutions.