Short-term EV Charging Load Forecasting Based on GA-GRU Model
Lei Guo, Peiran Shi, Yong Zhang, Zhengfeng Cao, Zhuping Liu, Bin Feng
Abstract
As a cleaner and environmentally friendly travel mode, the popularity of electric vehicles(EV) increased in recent years. The increasing charging load of EV will have a specific impact on the existing power grid. Different from the conventional load, the charging load of EV has great randomness. To accurately predict the changes in the charging load of electric vehicles, the K-means algorithm is used to cluster the charging curves of electric vehicles at each station. Then a gate recurrent unit (GRU) neural network predictive model is proposed, whose inputs include historical charging load power, weather data, and date types. Simultaneously, a genetic algorithm (GA) is used to optimize the hyperparameter selection of the GRU network, forming the GA-GRU model. Finally, it is verified that the model can effectively predict the short-term load of EV through calculation examples of the North China.