Digital Twin-Based Simulation of Electric VehicleInteraction within Smart City Energy Networks

Abstract

The integration of electric vehicles (EVs) in smart city energy networks creates substantial challenges for managing
energy, stabilizing the grid, and optimizing infrastructure. For this reason, this paper studies both AI and IoT (in conjunction
with digital twins) to manage wireless charging and Vehicle-to-Grid (V2G). We consider AI optimization methods that can
support predictive energy management and real-time decision making, such as reinforcement learning, genetic algorithms, and
deep learning. The recommended framework includes edge-computing architecture, federated learning techniques, and
Multiphysics models of EVs to improve grid stability and service quality. Our systematic review of the state of the art for
digital twins has revealed significant research gaps regarding integrating wireless power transfer and cybersecurity methods.
Our simulation results indicate significant efficiency improvements with peak load reductions of 18%, improvement in
forecasting accuracy of 29%, and reductions in state of charge error. In addition, we demonstrate 11%-15% reductions in
energy costs and an increase in battery lifespan of 8%-12%. Overall, this work provides a comprehensive roadmap for
developing intelligent, sustainable and resilient EV charging ecosystems that will resolve technical, economical, and ecological
considerations while protecting the privacy of data and securing the system.