Stochastic Modeling and Forecasting of Load Demand for Electric Bus Battery-Swap Station

Electric-vehicle (EV) battery-swap stations (BSSs) have become important infrastructures for the development of EVs to extend their driving range. Due to the randomness of batteries' swapping and charging patterns, the load demand of the BSS has a stochastic nature. It is necessary to investigate the charging load characteristics of BSS to guide the coordinated battery charging for mitigating the impact of disorderly charging behaviors on the distribution network. Under the uncontrolled swapping and charging scenario, four variables are essential: 1) hourly number of EVs for battery swapping; 2) the charging start time; 3) the travel distance; and 4) the charging duration. Taking these factors into account, a novel model based on Monte Carlo simulation is presented to estimate uncontrolled energy consumption of the BSS. Then, a generic nonparametric method for the estimation of prediction uncertainty of charging load demand is introduced. Adopting an actual typical BSS as an example, the simulation results show that the proposed prediction methods of the BSS charging load and probabilistic interval are suitable for forecasting the horizon 24 h ahead.