Energy Management of Electric Vehicles: AI-Driven Strategies for Smart Grid-Connected Charging Hubs

Abstract

This paper presents an innovative approach to improving electric vehicle (EV) routing in smart cities by combining heuristics and discrete-event simulation, specifically addressing the team orienteering problem. Initially, a biased-randomized heuristic approach is used to discover the most efficient EV routes, but this does not take into account the limitation imposed by the interconnected nature of the power grid. While these routes appear efficient when assessed independently, a more comprehensive evaluation is performed using a simulation model that includes the complex network of interconnected charging stations. This crucial second phase evaluates the practicality of the heuristic solutions in real-world scenarios by investigating how node interconnection affects route efficiency and charging station service. The simulation results provide concrete insights into optimizing EV mobility, enabling route adaptability to the dynamic urban landscape. This approach efficiently bridges the gap between theoretical optimization and actual implementation, emphasizing the need for simulation in verifying and refining routing strategies.