Data-driven intelligent EV charging operating with limited chargers considering the charging demand forecasting

Coordinated charging scheduling can improve the operating economics of charging stations and reduce the required amount of charging facilities. However, existing optimal scheduling schemes either simplify the charging station capacity modeling when taking into account the traffic uncertainty, or ignore the future charging demands when considering charging capacity limitation. To tackle this issue, a data-driven intelligent EV charging scheduling algorithm is proposed in this paper, by scheduling in response to the time-of-use (TOU) electricity price, the limitation of charging facilities, and detailed charger operating process is also considered. First, based on the neural network algorithm, a charging demand forecasting method is introduced to establish the charging task of the charging station. Then, according to the established task, an optimization model that considers the charging costs, battery degradation, and users' dissatisfaction comprehensively is proposed. The proposed model is formulated as a mixed-integer nonlinear programming problem, and a corresponding approach for solving the model is also proposed. Finally, the real-time operation process of the proposed scheduling method in the actual charging station is presented. By comparing with the existing methods, better effectiveness and performance of the proposed scheduling method are verified by simulation results.

Automated summary

A data-driven intelligent EV charging scheduling algorithm is proposed in this paper, which considers the charging costs, battery degradation, and users' dissatisfaction comprehensively. By forecasting the charging demand and establishing an optimization model based on time-of-use electricity price and charging facility limitation, the proposed algorithm achieves better effectiveness and performance compared to existing methods.