A Green IoT-Integrated AI-Based EV Scheduling Scheme for Efficient Charging Station Selection

The energy market has witnessed the drastic penetration of on-road electric vehicles (EVs), which serve as a robust and energy-efficient solution for mitigating the high cost, high energy consumption, and environmental pollution challenges of fossil-fuel vehicles. However, the huge energy demand of EVs needs to be coordinated for charging based on the available charging infrastructure, considering the critical factors of fluctuating prices and distance corresponding to the charging station (CS) and energy demand of the EVs. Therefore, we propose an Artificial Intelligence (AI)-based EV charging allocation scheme by performing the CS selection considering the EVs state of charge (SoC). We have applied the neural network model to predict the EV SoC to check their battery level for scheduling purpose. Further, we have formulated various CS selection scenarios based on the price and distance by assigning priority to the scenarios to enable efficient EV scheduling. Finally, the proposed scheme is evaluated considering various performance metrics such as loss curve, MAE curve, and SoC prediction plot in which adam optimizer yields the best result for the neural network model.