User-centric Charging Service Recommendation for Electric Vehicles

The rapid expansion of the Internal Combustion Engine (ICE) vehicles in the transportation system has caused serious concerns about air pollution, prompting the introduction of low-carbon vehicles as an alternative solution. However, establishing a transition to Electric Vehicles (EVs) demands appropriate infrastructure before large-scale manufacturing can be feasibly sustained. Unfortunately, the steep installation costs and the improper placement of Charging Stations (CS) have prevented the development of charging infrastructure, impeding its ability to satisfy the escalating number of EVs. EV manufacturers are trying to elevate user satisfaction and alleviate range anxiety by accurately forecasting State-of-Charge (SoC) amounts and offering user-centric suitable recommendations among all types of available charging services (stationary/fixed or mobile). Nevertheless, the scarcity of historical data for Artificial Intelligence (AI)-based predictions poses a significant difficulty. In the paper we propose an optimized recommendation system within the Internet of Vehicles (IoV) framework, considering both stationary and mobile charging services to optimize the overall experience for EV consumers and owners. The theoretical analysis and evaluations demonstrate a more efficient optimization of recommendations for EVs in designated areas, affirming the effectiveness of the recommended system.