Identifying Optimal EV Charging Station Locations: A Smart Grid and Machine Learning Approach

As the detrimental effects of fossil fuel consumption become increasingly evident, the quest for sustainable energy solutions has intensified. The transportation sector, a major contributor to global CO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> emissions, stands at the forefront of this transition. Electric Vehicles (EVs) have emerged as a key player in this shift, offering a substantial reduction in harmful emissions and a path away from fossil fuel dependence. However, widespread EV adoption faces significant challenges, including power grid stress as well as the insufficient and uneven distribution of charging stations. To address these issues, researchers have developed various models to improve the viability of EVs, focusing on extending driving range, enhancing battery life, and optimizing charging infrastructure. Innovative approaches have been introduced to strategically place charging stations and develop predictive models that recommend the best charging locations based on multiple factors. This paper introduces an innovative smart grid system for EVs that employs network clouding to facilitate seamless communication between vehicles and optimal charging station location. The system uses machine learning algorithms including linear, polynomial, random forest, and decision tree regressions to identify the most energy-efficient charging stations, considering parameters such as location, route, road conditions, weather, and traffic. Utilizing real-world datasets of the Volkswagen e-Golf, the proposed model aims to enhance efficiency and locate the optimized EV charging system.