At present, the most urgent problem to be solved in forecasting the charging load of electric bus fast charging station (FCS) is that the precision and accuracy of the forecasting models proposed by existing studies are not high enough. Complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) algorithm based on Gaussian white noise decomposes the historical charging load data into several intrinsic mode functions (IMF) and a residual component (Res) with different frequencies and complexities, and the sparrow search algorithm (SSA) is used to obtain the optimal control variables of the long short-term memory (LSTM) neural network. Based on the above two algorithms, CEEMDAN-SSA-LSTM model is constructed to forecast charging load of electric bus FCS. Taking the historical charging load data of Beijing Longjinyuan electric bus FCS as an example, the experiment shows that the prediction accuracy of the this combined model are greatly improved when comparing with the LSTM and CEEMDAN-LSTM model.
Day-ahead charging load forecasting of electric bus fast charging station based on CEEMDAN-SSALSTM
28.11.2023
1196439 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Electric vehicle charging station load calculation method
Europäisches Patentamt | 2020
|Electric Vehicle Charging Load Forecasting based on Monte Carlo
VDE-Verlag | 2022
|Charging load control method and device for electric vehicle charging station
Europäisches Patentamt | 2023
|