Title :
Research on electric vehicle charging station load forecasting
Author :
Xie Feixiang ; Huang Mei ; Zhang Weige ; Li Juan
Author_Institution :
Sch. of Electr. Eng., Beijing Jiaotong Univ., Beijing, China
Abstract :
In recent years, due to the pressure of energy crisis and environmental pollution, Electric Vehicle (EV) has gained opportunities for development. With the large-scale construction of charging station, the wide use of EV will cause the rapid growth of the power load in local areas. As the essential part of grid loads in the future, the charging station load forecasting, especially the short-term load forecasting, will play a very important role in production arrangement, economic dispatching, and safe operation of electric power system. The traditional power load forecasting model is mainly based on the factor of weather (such as temperature and humidity). Compared with the traditional power load, the EV charging station load is more complicated and mutable. In view of present EV charging station load, the trend of charging station load curve is more closely related to the user action and the flexible factors of charging rather than weather. Taking the distinctive characteristics of EV charging station load into consideration, an approach to accommodate this change by establishing the suitable model for the charging station load forecasting is presented in this paper. Based on the daily load data of Beijing Olympic Games EV Charging Station in 2010, this paper gives a brief introduction of characteristics of the charging station load and establishes three types of daily load forecasting model for EV charging station load, including BP neural network, RBF neural network and GM(l, 1) model. The application of the models has been realized in MATLAB.
Keywords :
air pollution control; backpropagation; battery powered vehicles; load dispatching; load forecasting; power grids; power markets; power system security; radial basis function networks; BP neural network; Beijing Olympic Games EV Charging Station; GM(1 1) model; Matlab; RBF neural network; charging station load curve; economic dispatching; electric power system safety; electric vehicle charging station; energy crisis; environmental pollution; power grid; power load forecasting model; production arrangement; Load modeling; Sun; Training; BP neural network; EV charging station load forecasting; GM(1,1) model; RBF neural network;
Conference_Titel :
Advanced Power System Automation and Protection (APAP), 2011 International Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4244-9622-8
DOI :
10.1109/APAP.2011.6180772