Load forecasting of electric vehicle charging station based on grey theory and neural network. (November 2021)

• Record Type: Journal Article
• Authors: Feng, Jiawei; Yang, Junyou; Li, Yunlu; Wang, Haixin; Ji, Huichao; Yang, Wanying; Wang, Kang
• Is Part Of: Energy reports. Volume 7(2021)Supplement 6
• Page Start: 487
• Page End: 492
• Publication Date: 2021-11
• Subjects: Load forecasting -- Charging station -- Grey theory -- Neural network; Power resources -- Periodicals; Energy industries -- Periodicals; Power resources; Periodicals; Electronic journals; 621.04205
• DOI: 10.1016/j.egyr.2021.08.015
• Languages: English
• ISSNs: 2352-4847

Abstract

Abstract: The rapid development of electric vehicles (EVs) makes the load of electric vehicle charging stations (EVCSs) affect the power grid. Aiming at the low accuracy of charging station load forecasting caused by the number of EVs, temperature and electricity price, and other factors, this paper proposes a load forecasting method of EVCSs based on a combination of multivariable residual correction grey model (EMGM) and long short-term memory (LSTM) network. Firstly, load influencing factors are analysed, and the grey theory is introduced into the load forecast of EVCSs. The role of EMGM in taking into account the effects of multiple factors and eliminating cumulative errors is analysed. Then, the EMGM and LSTM networks are combined to establish a mapping from the influencing factor data to the forecast, reducing the load forecast error of EVCs. Simulation and experimental results show that the accuracy of EVCSs' load forecasting can be improved by this method.