Abstract:

Electric vehicles (EVs) are emerging as the pillar of sustainable transportation, and efficient integration of EVs into Intelligent Transportation Systems (ITS) necessitates optimal charging strategies. Timely, cost-efficient, and grid-friendly EV charging is essential for the large-scale adoption of EVs. Nevertheless, current approaches tend to have high computational complexity, poor adaptability to real-time traffic and grid conditions, and suboptimal resource allocation, thus causing longer charging delays and energy imbalances. To mitigate these limitations, this research proposes a new framework named GA-EVChargeOpt, which applies Genetic Algorithms (GA) for EV charging schedule optimization in ITS settings. The method proposed here combines real-time traffic information, predicted energy demand, and availability of charging points in a multi-objective optimization framework. GA-EVChargeOpt is tailored to minimize total waiting time, alleviate peak energy demand, and improve energy efficiency by dynamically redistributing charging assignments in response to changing transportation and grid conditions. Experimental results illustrate that GA-EVChargeOpt considerably outperforms conventional heuristics and static scheduling. The suggested strategy leads to a 25% decrease in mean waiting time and a 30% increase in efficiency in the distribution of energy loads, reflecting its capability to promote EV charging infrastructure utilization as well as sustainable urban mobility.