Flexibility Probabilistic Prediction and Stochastic Optimal Dispatching of Electric Vehicle Clusters

Electric vehicles (EVs) are an important type of adjustable load resource, and predicting their flexibility is an essential prerequisite for implementing optimal dispatching and control. First, we propose a method to construct the cumulative energy envelope of EVs based on the data recorded by charging stations, thus forming a historical dataset reflecting the dispatch flexibility of EVs. After that, we propose a probabilistic prediction method of EV cluster flexibility based on Gaussian process regression for the high randomness of charging load. Finally, we establish a stochastic optimization method considering EV cluster satisfaction and propose a collaborative intra-cluster control method based on individual satisfaction. The simulation results show that the proposed methods can accurately predict the flexibility of EVs and then implement effective optimization and control of the EV cluster.