Clustering-based optimal operation of charging stations under high penetration of electric vehicles

Abstract

With the rapid growth of technologies related to electric vehicles (EVs), the market penetration of EVs has increased significantly in recent years and has resulted in an increase in the charging load demand for charging stations. In addition, the uncertainty in EVs’ schedules introduces various difficulties for optimal operation of charging stations. Operation of a charging station with consideration of individual EV behaviors may result in computational complexity and lead to infeasibility. Therefore, this study proposes an operational strategy to reduce the computational burden while maintaining EV schedules for charging stations under high penetration of EVs. The proposed method consists of two major stages, i.e., clustering of EVs and optimizing the operation of each cluster. In the first stage, EVs with similar attributes and behaviors are grouped into different clusters, where EVs in the same cluster share similar schedules and operational constraints. This significantly reduces the computational burden as well as the complexity of optimization modeling. In the second stage, instead of considering the behavior of hundreds of EVs, the optimal operation of each EV cluster is considered. In addition, the proposed method allows cooperative operation among EV clusters via vehicle-to-vehicle (V2V) services. The proposed cooperative operation strategy not only reduces the imported power from the utility grid during peak price intervals, but also significantly reduces the amount of energy storage required in the charging stations. This results in the minimization of the investment and operation costs of the entire system. Finally, the effectiveness of the proposed method is verified for an EV charging station with over 180 EVs.