A Real-Time V2G Scheduling Strategy Based on a Rolling Forecast-Scheduling Framework

With the increasing number of electric vehicles (EVs), electrical loads have risen significantly across most regions. To alleviate the pressure on the power grid caused by EV charging demand, load forecasting and vehicle-to-grid (V2G) scheduling have become active areas of research. However, uncertain conditions make it difficult to obtain accurate forecasted loads through day-ahead forecasting. Therefore, this paper proposes a rolling forecast-scheduling framework that integrates rolling load forecasting and updated EV charging information, aiming to provide up-to-date data for real-time V2G scheduling. First, a rolling forecast-update mechanism is proposed to utilize the most recent actual load values as input for forecasting the remaining loads of the current day. EV charging information is submitted only when EVs connect to the grid and is continuously updated throughout the scheduling process. Then, a rolling forecast-scheduling framework is proposed to ensure that each V2G decision executed at each time period is optimal. Finally, comparative experiments demonstrate that the proposed framework enhances the performance of real-time V2G scheduling by leveraging real-time updated load values and EV charging information.