Review on Scheduling, Clustering, and Forecasting Strategies for Controlling Electric Vehicle Charging: Challenges and Recommendations

Abstract

The usage and adoption of electric vehicles (EVs) have increased rapidly in the 21st century due to the shifting of the global energy demand away from fossil fuels. The market penetration of EVs brings new challenges to the usual operations of the power system. Uncontrolled EV charging impacts the local distribution grid in terms of its voltage profile, power loss, grid unbalance, and reduction of transformer life, as well as harmonic distortion. Multiple research studies have addressed these problems by proposing various EV charging control methods. This manuscript comprehensively reviews EV control charging strategies using real-world data. This review classifies the EV control charging strategies into scheduling, clustering, and forecasting strategies. The models of EV control charging strategies are highlighted to compare and evaluate the techniques used in EV charging, enabling the identification of the advantages and disadvantages of the different methods applied. A summary of the methods and techniques for these EV charging strategies is presented based on machine learning and probabilities approaches. This research paper indicates many factors and challenges in the development of EV charging control in next-generation smart grid applications and provides potential recommendations. A report on the guidelines for future studies on this research topic is provided to enhance the comparability of the various results and findings. Accordingly, all the highlighted insights of this paper serve to further the increasing effort towards the development of advanced EV charging methods and demand-side management (DSM) for future smart grid applications.

Keywords

• Electric vehicle charging
• scheduling
• clustering
• forecasting
• probabilities
• machine learning