Charging Load Forecasting of Electric Vehicles Based on VMD–SSA–SVR
Yanjuan Wu, Peizhi Cong, Yunliang Wang
Abstract
In order to reduce the impact of electric vehicles (EVs) charging load on the power grid, an EV load prediction model based on variational mode decomposition (VMD) and support vector regression (SVR) is proposed against the background of the real-time electricity price (RTEP) and the real-time ambient temperature (RTAT). Firstly, the historical charging data is decomposed into a series of modal functions with different characteristics using VMD algorithm. Furthermore, the decomposed data is combined with the RTEP and the RTAT, SVR is used to establish the prediction model, and the penalty factor C and kernel function parameter g of SVR are optimized using the Sparrow Search Algorithm (SSA). Finally, using the charging data of a charging station in a city in southern China as an example test verifies the effectiveness of the model.